• Karan Bhandari and Pooja CK discuss AI adoption barriers in development and security contexts.
• Pooja CK analyzes vendor data to improve business decisions and seeks insights on AI challenges.
• Karan highlights the need for multiple MCPs to connect custom data resources effectively.
• Current AI models lack explainability, complicating decision-making and ethical considerations.
• The integration of domain-specific agents could enhance collaboration and efficiency in development tasks.
• Trust frameworks are essential for guiding non-technical users in AI-driven decision-making processes.
• Data privacy concerns arise from sharing sensitive information with AI systems without proper safeguards.
• Continuous human oversight is necessary to ensure AI outputs align with organizational standards and ethics.

Karan Bhandari and Pooja CK discussed the challenges of AI adoption, emphasizing the need for improved explainability, trust frameworks, and continuous oversight to enhance decision-making in development and security contexts.

Transcript

Pooja CK: Yeah, cool.

Karan Bhandari: You Reached out to me to talk about the barriers to AI adoption. We’re going to treat this like an interview about my role as a developer and what we’re working on.

Karan Bhandari: We both work at Cisco. I’m a developer in the cybersecurity wing under the reporting layer of the Cisco umbrella. Would you like to introduce yourself, Pooja?

Pooja CK: I’m on the vendor management office team and I work as a data analyst.

Karan Bhandari: What project are you working on and what excites you about it?

Pooja CK: I handle vendor data to understand how our suppliers are performing in resource delivery and technology. I support the team with data and insights to improve decision-making. I’m also exploring AI and emerging technologies, and I’m part of a program to identify AI barriers and propose solutions. That’s why I wanted to talk with you.

Karan Bhandari: We both took part in a hackathon where we processed news and converted it into security policies.

Pooja CK: Yes, that hackathon was a great starting point.

Karan Bhandari: I’m glad we both enjoyed that experience.

Pooja CK: Shall we dive into questions? I know you’ve been in development for many years and you’ve tried various AI tools to boost productivity. What do you see as the biggest barriers to adopting AI solutions in a development context?

Karan Bhandari: We often start with generative AI via chat interfaces or notebook assistants. Developers then use IDE plugins like GitHub Copilot. We feed these models context—documents, source code, customer data—and expect useful output. But models can’t always give the best answers. When a model’s training cutoff is outdated, we rely on specialized connectors to pull in up-to-date code docs or custom data. Many vendor-provided connectors limit how many services you can hook together, so we turn to open source alternatives for more flexibility.

Karan Bhandari: Most developers only have client-side access, so they can’t fine-tune large models without GPUs. We end up using cloud services for training because local machines lack the needed hardware.

Karan Bhandari: Large models consume a lot of API resources per query. We need smaller, domain-specific models trained on our own stacks—Java, React, security logs—that work more efficiently for precise tasks.

Karan Bhandari: Beyond code generation, AI in neural network training still requires careful hyperparameter tuning and optimization. Low-code tools exist but often lack enterprise approvals. A multi-cloud strategy helps but adds operational complexity. Container orchestration systems are improving GPU support but it’s not yet seamless.

Pooja CK: Can you summarize your main point in one statement?

Karan Bhandari: We need more specialized connectors to custom data sources because without real context AI will hallucinate. Developers must act as analysts to frame requirements clearly, and we need cloud-agnostic training and inference that doesn’t lock us in.

Pooja CK: Do you use multiple clouds?

Karan Bhandari: You can spin up VMs in any cloud, but maintaining networks, security, and infrastructure is a heavy lift. Real-time learning is still difficult; we usually train offline and redeploy, so true online adaptation is rare.

Pooja CK: Do you see any AI models that don’t exist yet but would be useful?

Karan Bhandari: Most large models train on broad internet crawls. Many valuable data silos remain isolated. We need models fine-tuned on specialized internal datasets and robust ethical guardrails to prevent bias or harmful outputs. Security against adversarial inputs, like one-pixel attacks on images, is also critical.

Karan Bhandari: Explainability is another major gap. Decision-tree models can explain rationale, but neural networks are largely black boxes unless you use techniques like LIME. We need better built-in tools for transparent AI decisions.

Pooja CK: You’re right. Explainability is a clear pain point and something that urgently needs more work.

Karan Bhandari: When given precise instructions and test cases, AI can do a solid job. But if things change we must retrain the models. Human oversight remains essential to catch deviations.

Pooja CK: If you could remove one major limitation to make AI truly impactful for development, what would you choose?

Karan Bhandari: I’d build an ecosystem of collaborating AI agents—backend, frontend, DevOps, security—each an expert in its domain. They’d plan together, execute tasks, and hand off to each other, all under human supervision for validation.

Pooja CK: That collaborative agent framework would address many data and workflow challenges. Thank you for that insight.

Karan Bhandari: And after development, AI can also automate minor deployment tasks—preparing release branches, updating tickets, generating documentation—so us humans can focus on bigger problems.

Karan Bhandari: AI will still make fewer mistakes than humans overall, but we need security agents to run checks, enforce code quality, and prevent data leaks. Knowledge is the final frontier—defining workflows, gathering contextual data, and creating a central marketplace for reliable connectors.

Pooja CK: Building trust among non-technical users requires clear guardrails and frameworks. We must enforce domain constraints, structured outputs, ethical policies, and human-in-the-loop checks to protect sensitive data and prevent biased or harmful decisions.

Karan Bhandari: Exactly. Data cleansing, privacy controls, and consistent human oversight are the keys to safe, accountable AI adoption.

All Data Science Preprocessing steps

Refer to https://github.com/kurtzace/Diary2021/issues/5 for code snippets related to this podcast.

AML Intensity project (text preprocessing) https://github.com/kurtzace/AML-Intensity

Approximate Transcript

This is a technology racing podcast and you are listening to current Bhandari. In this episode we will be covering data science pre processing steps.To clean, manage and ensure that you have the perfect data for your machine learning inputs.At first we expect you to eyeball the data using either Excel or LibreOffice Calc or maybe using pandas. Take a look at how the how the data looks like. Check how data entry is done. For example, people tend to put entries like, no clue unknown, not available any. So take a look at the data entry patterns. Beyond that, take a look at the commands in the data like.Some people say it effort to this URL refer to that URL or effort to to do instructions like some people place to do items, pending items.And then take a look at some generic names like name, place, animal things. Try to see if you know you see odd things like numeric names or extremely vague country codes, or something like or something that defies logic. For example, if you see engines are becoming beyond V8 engines or V12 engines, if you see something like 48 or 99, just think of that.To be like a data entry error and some people put 99 are some in finite value. Some people put 999 or something. Something like in finite because they can’t write that inverted 8 symbol and then of course try your best to remove personal identifiable data so that you don’t get into GDP. Our issues look for name, age, SSN, date of birth, blood group.In order to load your datasets into your am, I recommend you to use pandas. Pandas is an open source library that can help you to read CSV, Excel files an.If you also wish to eyeballin.The online portals Google Big Query is a very good viewing agent as well. You can import files in Avro CSV. Jason an. You can view how the file looks like, but imagine if you are using pandas then you do PD import pandas as PD and PD dot read CSV. Take a look at type DF dot head to leak. Take a look at the columns or you can directly access by doing DF columns.But if when you do DF redhead it’ll show you the first five rows an the possible entries that have been there against it.You can also specify an upfront any values. For example, pandas in the read underscore CSV method accepts the not available value. So if your data entry has mentioned things beyond what the generic ones are, you can add them to the NA values. Then after you’ve loaded it before, you.You know, do any further processing. Take a look at df.info. That will give you account of not null values, null values, total rows count so that you know whether if it’s under 50,000 you can process that within one machine. If it goes beyond that, maybe you have to think of setting up, you know, using Google Bigquery Redshift.Or data SQL, or as your data lake. Or maybe set up a multi machine environment with spark. But let’s not get into it. Imagine that if you know you have reasonable number of rows, you can type DF dot esnal to see how many numbers are there and you can do DF traditional dot some to see what is the total count of the nulls.Andosometimes some people have the habit of putting a zero against things like glucose, blood pressure, skin thickness. You know that you know these things cannot be 0 or BMI. You can go ahead and treat them like null. So what you can do is you can try it. DF put the list of all the columns and replace zero with NP dot NNNP comes from NUM PY.When you import number.And then take a look at.You know the domain knowledge, I mean based on your domain knowledge, like us as I had mentioned, the engine count or the weights of people who age of people. You know that it will be within a specific range.Type DDF dot describe data frame that describe and you’ll be able to see the mean, median, mode, minimum. Where does most of the range lie under the data? Look normalized to you.And then some, and you have to, you know, sometimes standardized data for.Pre 44 before you feed it into a male model, because many things appear to be different. For example, number of bedrooms maybe 234, but your area will be.Like 5000 square feet, 2000 square feet, and I think both are different, but the machine learning algorithm will feel that you know the one that size represents appears to be more significant because of the kind of value it is having. So try to use SK loans, preprocessing min, Max scaler. You can give an arranged OK between zero to one. Run a fit transform on the Min Max scaler and you will be able to.Standardize the values between zero and one, so your number of bedrooms will appear between the range of zero to 1, even though it may be 10 bedrooms and even the price or the square feet can be normalized into a value between zero to one. You could also use, you know the standard scaler instead of the min, Max scaler and standard scaler also has similar type of output.Sometimes you know your data set may have same meanings, so various columns may talk about the same kind of meaning. For example, I think in the complete data science course they had mentioned that.When religion voting history participating in associations are bringing, maybe they reflect your attitude towards immigration so you can merge them into a single column and then sometimes you are analyzing things that are related to housing, but some space information about particle information comes into your data and you know that they are not correlated or something like your blood parimeter comes in into your.Data set or maybe even ID column. You know the idea of SQL does not give any effect to your data science model. It may just confuse it further and treat the ID to be like a significant factor in.Predicting the outcome so you can remove uncorrelated data an.That way you will ensure that you know it’s not creeping and you can also study is so. So if if you’re if you’re doing something like a linear regression or logistic regression, if you use statsmodel, you’ll be able to study the P values and even the F statistiques. So if the P value is.Around 0.050, then it’s considered that you know the.Data set is significant, so there is also something called as.The variance inflation factor, so that also comes from statsmodel. I’ll talk about that very soon.I think I’m jumping the gun here, but yeah, let’s go back to data whitening an then we will go on and head towards to the P values and.The stats related information.Sometimes you may have to convert if you have unique values between 2:00 to 10:00. For example, if there are a few categories that OK, this is non veg veg.Order something like carnivorous herbivorous or omnivorous. There are three categories, so you can convert the small number of categories to one hot encoded format and it will be represented like a binary number, like 001.110 or something like that so.For that you can use PD dot get dummies to use the one hot encoding. Or you could use SK learns preprocessing one hot encoder.An there is also something Called’s label encoder where you can mention.The rank of the category. For example, if you feel you know gold has a better rank or silver, has a medium rank and bronze has third rank, you can represent them to be like a category of 123.Out, so it’s not recommended that you use gold, platinum, silver as machine learning inputs. You there is also one more lesser used option called Bina Riser which is given by SK Learn, but that’s not used very often.Sometimes if you have large ranges then how do you deal with it? Maybe you could take into account for example BMI values may be spread out over a / a range like 20 to 40. Then you could claim like OK of BMI less than 18 is underweight and maybe between 18 to 25 normal and 2230 years overweight and 30 to 40 is obese so you can take that approach an.Try to reduce your categories. You can use DF dot log to locs to you know. Take a look at what the ranges are and create new data columns to.Do.Sure that you know. Do you have reduced the number of categories?Now before I get into image preprocessing.I would like to talk about other things. For example, there are a lot of statistics variables that you may have to acquaint yourself with that.Off, for example, there is mean which is almost like the middle part, and then there is the median where most of your data set is lying under. So for example, the median height of most of the world is between.5 foot to 6 foot.An if they are adults an and you have to find out what’s the median and mode is like the most frequently occurring one, and then there is something called as variance inflation factor. That is, it estimates how much is the variance of coefficient is inflated because of linear dependence with other predictors. For example, it tells.I think if you take a look at a VIF of one, you need to study.So so anyway, so VIF under 10, if we IS is under under 10, it’s borderline, but if VIF is between one to five, it’s perfectly OK.And maybe if you’re VIF is exceeding 10, that’s a cause of concern. For example, if you see too much variance in the mileage or the engine capacity, then you know we are F comes into picture so.I saw it. It shows it talks about situations where you know you have an anomaly in the data set. You can use a VIF.So just try to see if your VIF is under 125. You can use the stats model VIF.So you can see which one is an outlier. For example, you may find that you know certain extremely skilled people like Michael Jordan is skewing your basketball skill measurement.Anne, and sometimes you have extremely odd values because of OK. One extremely rich individual purchase the entire city. I mean these are outliers and it’s not very generic. But then your entire analysis gets skewed by a very large margin.And on some it’s good to create checkpoints while do you process your data because you’ll be dropping. You can also do DF drop to drop your unnecessary data columns, but I feel that make a copy like DF copy can make a new copy of your data set before you do something very different so you can come back to these checkpoints. And of course I forgot to tell you like just take a look at some unique values to see.RDF dot Uni can help you to judge certain criteria that I had mentioned in the past.And grouping, I mean working with time is tricky.So you have to deal with different formats, so it’ll be good if you use a PD .2 datetime an try to see if you can do that and you can create groups also by using lock variable so you can see that OK columns 1/2/10 belongs to Group One. Columns 2/2/15 belongs to all columns, 15 to 20 belongs to group 2.So you can convert them to groups if you feel there are too many one hot encoded values.An yeah you can just see if you can use NP dot where to see if you know your.Data is exceeding the dot median values an. That way you can see if you know certain things are exceeding the median. You can have a new column like that.I know some people can even balance the data set by using.By using techniques like averaging by using basic averaging an. If you have to apply a function to every intro to every entry. For example, imagine if you are going through a list of companies and you also want a description about the company from Yahoo API. Then what you can do is you can do DF the column name and do dot apply and then you can use the pan.Do pythons request module to call Yahoo Information API and get one line description about every company? Or maybe the current stock price? So that way you can you know preprocess or data set or or get the ranking of the company so that you know certain things to augment your data set is very healthy.I’m.So I was talking something about the P value and P value is nothing but.You know you have to see what is causing evidence for your hypothesis testing in the sense it’s it’s used to check your null hypothesis value out the alternate ones, or to challenge the status quo.And.If your P values.If your P values are very, very high.Then you need to take a look at.8 versus the lower P value.As I had mentioned, the P value can assist you with finding out if something is adding noise to your data set. For example, if I just generate a random value an add a new column an it’s adding a lot of noise to your data set. So for P value is very high. For example, if the value is 0.1 or even or even 10, then that means that kind of high. The kind of hypothesis that you’re seeing that this value is actually helping your linear or logistic.Audio roomno. Machine learning model.It can actually give a prediction whether it’s really helping or not. So if you have a P value which is very low, that is less than 0.05. That means you’re that means the column is extremely helpful to predict, so this statsmodel can give your P value and then there is one F regressions package that comes with escalon that can help you with the P value to find out if a particular column is significant in helping you to predict. So if something which is uncorrelated and.An if you feel that you can drop it if the P value is coming to be 0.1, then you know you can reject your hypothesis that you know this column is actually helping you to do something very significant.Now, with respect to images, some images are slightly are extremely different. Actually, it’s you have to treat it in a different way. Sometimes if you’re dealing with documents, it’s better to run an OCR on top of it with the help of Microsoft Cognitive services or a WS text, extract an or maybe Google’s or GCP offering of OCR. The vision offering to you know do the OCR.Can help and then you can perform any are on top of it so when you do any are. Let’s talk about anywhere after the image section is done.I will talk about I’ll come to that very soon. So when you get images, you have to. You know sometimes fix the alignments, fix pixel values, maybe make it into grayscale.You need to ensure that you know they’re not very transformed, or maybe you can convert to black and white or convert to grayscale.Anne.Sometimes you’re OK with just the edges of the images, so if you have too much noise, for example, if you scan the entire ID card, I don’t think your object detection algorithms are powerful enough to do, you know, differentiate between what is the name, what is the address? So sometimes you’re just interested in, you know, extracting the face part of it. Then you could, you know, help it with the removal of certain features that look different, so you can use the histogram module to compare.Images features.And then you can use open CV to resize them, open them, view them, transform them.Or do basic things and you know do edge detection.Do corner detection.An there are a lot of image preprocessing techniques like you know removal of noise. I think resizing is very important otherwise.Young, you may be running out of RAM, and when it comes to text NL TK toolkit worth 2 VAC.I’m explode. They are very effective to deal with.Any us?So any artist named entity recognition an it’s nothing, but I’m using the term. Any are but actually what I’m talking about is text processing because.You can analyze everything. So suppose if you were asked to work on, you know some news data set then and if you asked to do something like find out the most significant topic in it or find out the most significant variable, then you know you need to remove punctuation’s remove stop words like A and remove certain things that.Good cause.Ambiguity, for example, some people do something called as stemming and lemmatization. So lemmatization is. You can for example you have good, better, best or amazing. You can convert it to something like good. So then you have lesser words to deal with. If you do stemming and lemmatization will convert super later form.Will convert the Super letter form like extremely good or something like.Bravest Orpheus is to something like only fear so brave that’s the help of the stemmer.You could all stitch this into one pipeline and preprocess your text and that can help you know to reduce the amount of text that is going in. An even maybe could help you with the frequency counting.There is a very good library called his work to work that can help you to distinguish.Word related Dome.Matching or correlation. For example, you know that King and Queen are similar to each other, but there is a difference of a male and a female. So if you do King minus queen.It may give rise to mail, so you can do such kind of arithmetic over subtraction addition with words so you know we know that you know band and restriction is very close to embargo or.Things like computer and mouse.Are closely related compared to something very different like God, so you can use Google Word, two VEC preloaded models to help you to get the similarity score of one word versus the other, so that can help you to do word counting. For example, if you see the show notes, there is a link to a project called Aml intensity and I’m counting the intensity of of embargo blacklist.OFAC orderuh.The sanction on a particular company and that’s helping to calculate the intensity. Take a look at the project and you will be able to perform certain tasks related to text and the most important is any are the entire concept of any artist named entity recognition, so you can find out which one looks like a company, which one looks like.Date which one looks like a number, which one looks like a noun pronoun. Parts of speech. So that’s called parts of speech tagging. You can use space. He library to help you to do that.Image text data preprocessing is an extremely vast field. You may, as a data analyst, spend most of your time doing that.Because it’s real world data is not very clean. Of course, if you download some of the popularly available datasets like MNIST.Or diabetics one who data they are cleaned. Or by the Census Bureau.They’re extremely clean, but real world data is not that clean. I think you may need to spend some time to put your head around it, clean it, process it, and you may have to revisit it. It’s not just you, do it one time and you forget about it. You have to keep persisting it, save it and NPZ format, Civitan CSV format, save it in Excel format. After you’ve done it and maybe even save the intermediate steps you have to revisit it.So I wish you all the best in your data pre processing tasks. This is current binary signing out.I’m working as a developer in society general. You could contact me on Twitter or on LinkedIn.I’m known as KURTZAC codesys on these platforms goodbye.

Cheap ways of going to production

1. Use api gateways to invoke your lamdas

https://aws.amazon.com/api-gateway/pricing/

https://aws.amazon.com/lambda/pricing/

https://azure.microsoft.com/en-in/pricing/details/functions/

2. Use AWS codebuild for 2+ minutes of processing or build in azure dev ops

3. Complicated workflow? AWS step functions or Azure logic apps

https://aws.amazon.com/step-functions/pricing/

4. Use azure app function to host angular react application

5. AWS S3 or Blob azure for storage needs

6. google Colabs notebook for free GPU

7. Ms cognitive services/textract

8. if Load balances like elasticloadbalancing costing 22 USD is expensive, install consul as loadblancer in free azure web application -(https://github.com/Infinite-demo/Portfolio-Manager-micro-service-project) and corresponding youtube video

or simple let api gatway do load-balancing for you. https://dashbird.io/blog/can-api-gateway-act-load-balancer/

9. Analytics – azure application insights – 5gb is free. Or Cloudwatch of aws. Managed ELK is expensive

10. Dynamodb for DB needs – nosql

11. Notification hub from azure or AWS pin point or AWS SES/SNS

12. Create graphic assets using GIMP, Inkscape.

13. edit your product launch videos using openshot / camstudio – or create advanced animations in blender.

Pricing Overview

1. Microsoft Azure

Microsoft Azure Estimate
Your Estimate
Service type	Custom name	Region	Description	Estimated Cost
App Service(dotnet/java/python/node angular/react)		South India	Free Tier; 1 F1 (0 Core(s), 1 GB RAM, 1 GB Storage) x 744 Hours; Windows OS	$0.00
Storage Accounts		East US	Block Blob Storage, Blob Storage, LRS Redundancy, Hot Access Tier, 1 GB Capacity, 500 Write operations, 1,500 List and Create Container Operations, 1,500 Read operations, 1,00,000 Archive High Priority Read, 0 Other operations. 1,000 GB Data Retrieval, 1,000 GB Archive High Priority Retrieval, 1,000 GB Data Write	$0.03
https://dashbird.io/blog/can-api-gateway-act-load-balancer/		East US	Free tier	$0.00
Key Vault		East US	0 operations/mo, 0 advanced operations/mo, 0 renewals/mo, 0 protected keys/mo 0 advanced protected keys/mo	$0.00
Azure Active Directory		West US	Free tier, per-authentication MFA billing model, 10 MFA user(s), 25001-100000 directory objects, 0 Hours	$0.00
Azure DevOps			5 Basic plan licence users, 0 Basic + Test plan licence users, Free tier – 1 Microsoft Hosted Pipelines, 1 Self Hosted Pipelines, 0 GB Artifacts, 0 VUMs	$0.00
App Configuration			App Configuration is free during Public Preview.	$0.00
Cognitive Services		West US	Text Analytics: Free size, 5,000 included transactions with 0 overages.	$0.00
Event Grid		West US 2	1,00,000 operations per month	$0.00
Azure IoT Hub		East US	Basic Tier, B1: Unlimited devices, 4,00,000 msgs/day, $10.00/mo, 0 undefined	$0.00
Azure Open Datasets			There is no charge to use Azure Open Datasets during preview.	$0.00
Machine Learning Studio		South Central US	Free	$0.00
Azure Search		East US	Free tier	$0.00
Azure Bot Service		West US	Free Tier	$0.00
Support			Support	$0.00
			Licensing Program	Microsoft Online Services Agreement
			Monthly Total	$0.03
			Annual Total	$0.38

2. Amazon AWS

Service Type	Components	Region	Component Price	Service Price
Amazon S3 Service (US East (N. Virginia))				$0.14
	S3 Standard Storage:	US East (N. Virginia)	$0.12
	S3 Standard Put Requests:	US East (N. Virginia)	$0.01
	S3 Standard Other Requests:	US East (N. Virginia)	$0.01
Amazon DynamoDB Service (US East (N. Virginia))				$0
	On-demand Capacity:	US East (N. Virginia)	$0
	Provisioned Capacity:	US East (N. Virginia)	$0
	Indexed Data Storage:	US East (N. Virginia)	$0
	DynamoDB Streams:	US East (N. Virginia)	$0
	On-demand backup:	US East (N. Virginia)	$0
	Continuous backup (PITR):	US East (N. Virginia)	$0
	Restoring a table:	US East (N. Virginia)	$0
Amazon CloudWatch Service (US East (N. Virginia))				$0
	Standard Alarms:	US East (N. Virginia)	$0
Amazon SNS Service (US East (N. Virginia))				$0
	Requests:	US East (N. Virginia)	$0
	Notifications:	US East (N. Virginia)	$0
Amazon SQS Service (US East (N. Virginia))				$0
	Requests – Standard Queue:	US East (N. Virginia)	$0
	Requests – FIFO Queue:	US East (N. Virginia)	$0
Amazon SWF Service (US East (N. Virginia))				$0
	Executions:	US East (N. Virginia)	$0
	Chargeable Actions:	US East (N. Virginia)	$0
	Workflow Days:	US East (N. Virginia)	$0
AWS Data Transfer In				$0
	US East (N. Virginia) Region:	Global	$0
AWS Data Transfer Out				$1.26
	US East (N. Virginia) Region:	Global	$1.26
AWS Support (Basic)				$0
LAMDA	First 1M requests per month are free.		$0
		Total Monthly Payment:		$1.4
Also See	https://aws.amazon.com/free

3. Google Cloud

1. https://colab.research.google.com

(it is public though)

but you can get GPU/CPUs for google compute engine for free

Colaboratory is a free Jupyter notebook environment that requires no setup and runs entirely in the cloud.

With Colaboratory you can write and execute code, save and share your analyses, and access powerful computing resources, all for free from your browser.

4. CI/CD

a. Github/bitbuck/AWS codecommit allows 5 free users for private repository.

b. azure-devops-services allows

See https://azure.microsoft.com/en-in/pricing/details/devops/azure-devops-services/

1 Free Microsoft-hosted CI/CD
1 Free Self-Hosted CI/CD

Nuget/npm/mavin 2gb is free

Below is for 5 users in free plan

• 1 hosted job with 1,800 minutes per month for CI/CD and 1 self-hosted job with unlimited minutes per month

• Industry-leading NuGet Server

• Support for Maven, npm and Python packages

• Upstream sources to help protect open-source dependancies

• Integrated with Azure Pipelines

• Sophisticated access controls

 Azure Pipelines: Includes the free offer from INDIVIDUAL SERVICES

 Azure Boards: Work item tracking and Kanban boards

 Azure Repos: Unlimited private Git repos

•  Azure Artifacts: 2 GB free

4. ML workloads free notebooks

https://medium.com/@siddesh.001/top-5-online-free-notebook-ipynb-and-other-cloud-services-dbf9580d99e3

Approximate Transcript

This is episode 19 of the technology icing podcast and you are listening to Karan Bhandari working as a technical leader, associated generally, and in this episode I will be covering cheap ways of going to production. That is how to use the cloud for almost free and productionized your idea or your next big startup fling in a very affordable manner so every product that I will be talking about will be under $1. I will not.Consider any product that exceeds that threshold. Of course you need to use it very judiciously so.What first we’re going to production in are cheap and effective manner. Is the API gateway that calls your Lambda functions. So instead of using the traditional virtual machine or the EC2 machine, I recommend that you just write your function. What you generally put in our Web API controller or a spring rest controller and you call it.From a Lambda an what is alambra? It’s it’s offering available and given by AWS to just invoke your function and you need not worry about.The infrastructure, so in the modern Parlett it’s known as the.So with this way of computing the equivalent in Azure is called the Azure functions. Now there are various ways you can invoke Lambda. First is by making a web request, or you could have Cron job triggering it, or you could have something else, like a webhook, commencing it. So I think most people exposed web applications, so it will be good if you.Make the structure of your API using an API gateway where you define what is your port call. What is your post call and then let your API gateway connector to the Lambda function.So and then in the Lambda you take the payload and you return the response. Now all this is OK if your Lambda is going to take under 3 minutes. But what if you want to start a long running job? Well then I can’t use your Lambda because they have a restriction of two to three minutes. So in that case for slightly longer jobs I would recommend you to use a WS code build. I know it’s not used typically for processing your compute loads or you know passing your Excel and going through 1,000,000 lines.But I’ve seen code built gives you a command line interface, AWS code build, and that command line interface. You can run anything you want. You can run a shell script. You can run a Python script from the Shell and they give you various hooks like Preinstall Post install after install publish and inside inside that you can run your long running logican.Think you just have to pay 1 cent for every minute of computation.So what about your complicated workflows? If you are planning to have step one, Step 2, step three. OK. If suppose you get the money in from your accounting system, then you want someone to approve it. So there’s a workflow in it. If it’s under a certain dollar amount, it’s auto approved or maybe.If my management hierarchy air is responsible for this limit and then about them to be as responsible with some other limits. So if you have a complicated workflow, I recommend you to use a WS step function or as your logic applications so it has good integrations with Twitter outlook.And as your cognitive services.So you can run your complicated workflows using a double step function or as your logic apps. Now if you’re having a front end angular application, you can either put it in S3 bucket.Or you could, you know generate a friendly URL with HTTPS using.A day Azure app function, so if you host an application under 1GB which you will be able to host it as a free app. But of course you won’t get a domain name for it, that’s OK, we just wanted, so you’ll get a website likemyapplication.azurewebsites.net.And if you want for your machine learning training needs of free GPU, I recommend you to use Google Colabs. They give you a Jupiter notebook environment and there you can run your tensorflow or keras machine learning models and that will allow you to use the free GPU available and it will allocate a machine whenever you execute one of the cells of the Jupiter Notebook.Then if you want to process receipts invoices.Maybe ID cards address proof. You could use the Microsoft Computer Vision Cognitive service, or extract a WS2.I process your applications images for free for a very low rate, not free.And if you take some.If you take some load balancer which is ELB or the Equal International, it will typically cost $22.00. So what you could do is you could modify API gateway or little bit.To make it behave like a load balancer for you, I’ve posted a link in the article on how someone had managed to make the API gateway behave like a load balancer. Or you could install console. It’s a product of hashicorp, which incidentally does service discovery, but it can behave as a load balancer.If you put it with, if you have an ocelot API gateway written so there is a guy called Infinity demo who has made one proof of concept of making a stock manager application load balance using console, I’m going to put the link in the show notes and if you need as your insights to help you with Analytics you could make.You could make it.By using Azure application insights or cloud watch of AWS, the manage LK instance is expensive. So in Azure application insights you could dump file GB of worth of data an what data could it be OK if you’re using the front end, what the user clicking, what interactions are happening or you just want to analyze in Excel sheet or you want to make pretty graphs, sort of it, or you want to find out.How the sales are doing? Then you can use analytics for free using Azure application insights. Theyhave.net, react, angular APIs and cloud watch has rich integration with even the.With the command line.So you could use analytics for almost free, then very enough for a very affordable databases Dynamo DB.Which is so if you take an Azure SQL, it’ll cost you $6. Why use that when you can use Dynamo DB? I think for up to 24 up to five GB it’s under under a dollar and it is a no SQL like interface. The Mongo DB instructions don’t work out of the box with it, but the instruction set of it is very similar to mongo DB. But there is a higher degree of learning curve because you need to understand what is a read capacity on it. Right capacity unit, but it does the job.It’s a good database.Now for notifications.I I see a lot of services available online. For example there’s MailChimp. We can send an email or a WSS CS or SNS. Now I have seen there is one limitation in that is.Unless you send it from a domain, you will have a limit of 1 email per minute or at the Max 200 emails. So I would recommend that you get a very very cheap domain to send as many emails as you can.But if your limits if your limit is under 200 emails, I think SES is a very good option, and then there is notification hub from Azure or and there is sendgrid and I think 10 grid has a process they you need to give them a business use case and only if you fit their ecosystem they will approve you because emails are something which is very exploited by.On people to send spam so people are cautious about letting you use their emails for free because free meals can cause her work.Or it could cause the reputation of the product that you’re using to go down if used in not a very judicious manner. So yeah, you have to submit use cases to either AWS for either using the pinpoint service which is similar to MailChimp or using or if you have to unlock a higher limit in MailChimp.Are they also offer API integration as well?Even if you’re creating graphical assets, you could use jimp or Inkscape. Jim is similar to photo shop, but with with a slight modification and it’s open source. Inkscape is similar to code.But it’s free of course, and you can make good vector arts using it and you could edit your product, launch videos using openshot, record your screen using camstudio or create advanced animations using blender.So I hope.That is no hindrance to you for going to production. If you are a college kid, you could still make an enterprise grade application.Annyet compete against the best.So.Don’t let any barrier of how to go to prod Imperial growth.Let your wings free. Goodbye this is current bhandari. I’m working as a technology analyst and Associatedgenerallyasa.net and angular react developer also expertised in containerized offering such as docker EKS and kubernetes.This is me signing off goodbye.

My experiments with Openshift

 

Contents
What?
Advantages
Create dotnet
Create Nodejs
Create angular/react
Create python
Large model
Custom starter project
About java ecosystem
Tools
Secrets/Config maps/env variables
Resource limits
volumes
Dockerfile
Image deployment
tinker with kube side
rsh ssh
Courses

Links

https://www.youtube.com/channel/UCGngiDaHqxdat2aotHm-bDA

Free course https://www.udemy.com/course/deploying-containerized-applications-technical-overview/

https://www.tutorialspoint.com/openshift/index.htm

Approximate Transcript

Part 1

This is a Technology Icing podcast and you are listening to Karan Bhandari and in this episode we will be covering my experiments with openshift. Why open shift? I will show you? How do you deploy an application?In an container world without actually knowing what is a container, what is a docker and what is Cuban at ease?Open shift is away too abstract, even what is the underlying infrastructure you can be cloud agnostic. You did not know what is GDP? The GCE? Sorry the EC2 and as your VM or whether you are using any other underlying machine from maybe any other cloud provider like electrify.You can be cloud agnostic completely by using openshift. So open shift.I’m going to explain you from a user’s perspective, not as an operations person, so someone who is an application writer.And someone who just wants to put their workloads onto a container without actually knowing what a container is. I’m also going to talk to you about how if you were an operations guy, how do you modify the container to make it more amenable for your applications to be up and running in.The same Cuban 80s environment, an also if needed. You can modify the underlying yamil files of Cuban 80s if you wish to get into the nitty gritty’s and The Dirty parts.So these are the advantages you have ready made templatesavailableof.net, node, JS, Python, react, angular and Java ecosystem starter projects.All your concepts can map easily to.Open shift if you know communities and if you do not know.It’s OK, you can still be onboarded so you go to the GY.You click, you go to the catalog and want to deploy dot net application. I just go to dot net, add to cart an.I just give it to the GitHub Location an it will deploy it for me. It will give me a pod running with the service and I if I click on create route and exposes service it will do that for me. It will go ahead an just give me an end point without me knowing what is underlying machinery. What is a container? What is apart and there you go. A.net application is ready now if I need to give it a different starter project. For example if it’s not there in the route.Then I give it an environmentvariablelike.net speed underscore starter project and I tell it’s under source slash Myproject Slash Service Slash Project Dot CSS Praj an if I need to give it another nougat location, you can either commit a new Newgate dot config file within GitHub, or you can give it as an environment variable, and if your new gate is using a custom certificate that is not see S certified.You can give it another wreck.Directorywithin.net to contain your CSS words.Anosimilarlynotjust.net you can also deploy a node JS application. Most people typically right as a node, JS Express application and there is one more side which rights angle and react application. Now there is no template. If you go to Openshift catalog and search for an angular react will not get one.Some people use the NPX equal system and use a project called node shift. So in the root directory of the angular react application they typed spaced node shift an the name of the disk directory and it will you know do the deployment an give you an end point. But some people will also create an express index dot JS file and use a middleware called.Used Arctic so which is used to deliver the public directory. But you can make it point to the disk slash or application folder.If you are using wooden express, that’s a clever way to deploy an angular or react application.And you can also deploy a Python application by going to the catalog and you know make sure your GitHub location has a requirements dot TXT file, an app dot PY file.If your starter project is not app dot pie, or maybe something like underscore, underscore, net dot pie, you could tell it in an environment variable and even your pip configurations like your paper index URL or your pip.SSL checks could be configured from environment variables.Whatever I’ve seen, I’m seeing that you could do in the GY. You can also do in the command line. Now, in the GY you create something called a projects or many of projects, and one project is a collection of numerous applications. For example, if you’re in a library, the library and will have a set of applications for itself. A set of applications for the accountants, a set of applications for the end users. So it’s like a namespace that categorizes your your set of applications into numerous apps, so in.Or in Openshift, you create a new project and then you create an application within it.Now there are some challenges that come with you want to go away from what is considered like, uh, like you know a typical project like a Java project. Some people have some patterns that OK the that they will have to have a rabbit MQ running with Mongo DB running an A set of Micro Services communicating with each other. For some of them have some monitoring solutions like Cabana Prometheus.So they want to make their own template. You can create a template with the you know, like an orchestration you know orchestrate multiple pods.Or you can also create a project that is not very typical of being a.netbuta.net, which steroids then what you can do is you can make a custom STY image. Now an STI image enables you to make a similar.Type of application that is available in the catalog.And but you know, your developers can use that directly instead of using the default one that is generated for you. So for example, soon in an STI image you need to download the STI command line. In this you need to have Docker knowledge. What you will also create is set of executable scripts like will create a build script and assemble script and run script. So it will Anna Dockerfile and another assemble it. Will you have to tell it how to.Assemble all the binaries into one cohesive unit and the dockerfile is the one that you know pulled your dependency. Does your Yum installer app get and?In your runtime, you need to tell ok.net space. Run or maybe something like Docker compose up so you can write your own runtime instructions in a custom STY image and this can be available as one more image that the developers can use, so they will give something like.Use that image and use this GitHub URL so the GitHub URL will belong to the developer the the new image that you created from S2 I will belong to the operations team and you can separate the operations from the developers by making this custom arrangement.I’ve also seen that in Python people struggle with the deploying their AI solutions. With large models you can’t put a massive file onto github.com and just assume that it will work. Some of them use a new concept called as git, LF switches agate large file service. I don’t think you can directly use a large file system as is if you use a Python template, but you can get around by writing this application specific code. For example, use a bow to.Library to download a model from S3. Once the application starts and then you can mention that OK, my board is not ready with a separate readiness URL now I have mentioned that OK, you know the.Or just the superficial way of creating an application. But once an application is created, you can also go to the UI and click on the pod and you can click on edit Hamill and what you see underlying isn’t exactly abilities jamall file that has been generated for you in the background an you can toy around with it, tinker with it. You know you can play with it, you can change resource limits for storage that is a separate module.You cannot directly create storage on the UI an associated with your container is some path an it will map to the PVC concept of the provision volume concept of Cuban 80s within it.AnnThis way you can work with.Some applications that need storage.And I will talk about.How to even debug your containers? You could use RSH.And this other search will help you to SSH into the pod and debug what’s going on.So get cracking with it.Now in the dockerfile that you create, if you want to create a dockerfile then communities will allow you to do it and you can tell Cuban it is to use a dockerfile strategy. Just be aware that you cannot use any other.Um?Non Docker hub image as a starting project.If you.Want to use a custom image that is there in a private repository? You can directly make the image and push it to the Docker Registry private Docker Registry that is given by default to your project that has been created. Or you can also you know.Click you know create or.Oh OC secret aura which Maps to community secrets and makes it point to that private registry. But you cannot make a dockerfile out of it. I’ve had trouble doing that, but you can get around it with by using the concept of already premade image from an existing private registry.Know what you do on the on the graphical user interface can also be done with the command line and the command line directly Maps to Cuban 80s concepts or you can also just do oversee space, new app and the GitHub project name after you do OC. Login. That login command can be copied from your user profile section of the Geoeye.

 

Part 2

 

As mentioned, openshift can also work with Dockerfiles, an in this docker files. I think it’s mandatory for you to use an existing image from Docker hub an if for some reason you want to extend the existing functionality of.net or node JS. Then there are some specific openshift.net starter.Images that you can begin with.AnnI have. I’ve had trouble with deploying any application that had.An image which is the starting point of a dockerfile that is not a part of Docker hub. So for example, if I made a custom image in a private registry and I’m extending that in the dockerfile, I’ve had trouble with doing that. So what I generally do is I build it and put it into the private registry of Openshift. So when you make a new openshift project or default part of that is giving you a private Docker registry an.You can build your existing dockerfile.And then diploid diploid as an image. So you can directly deploye or Docker image.It could be from a private registry like AWSECR or something like the Azure Container Registry an or maybe your your custom one. You just need to link your secrets like you do OC. Create secretan you tell it it’s a Docker Secret and then open shift will be able to deploy the image and you mentioned the port that you want to deploy. An open shift will be able to do that for you. That’s for those users who want to get around the limitation of using a docker.Hub one, but yeah with I know that and an image from a private registry can be deployed as is into their into their registry. Or it could directly be diploid from the private registry if you give it the secret file for the Docker.An one disadvantage with open shifters. Or maybe it’s a good security feature, but I’m looking at it as a disadvantage is none of your containers will have root access, so you need to really right application. Well to ensure that you know your it’s not writing into random directories that are not declared in your configuration, like for example in Openshift you can create a volume on the Geo. Why you just create a volume and you say I want 1GB.Oh, like you create a storage an that will use the Docker and communities concept of PwC.Anne will link the volume to your container and in that you can write and in that the openshift will create a dummy user an accordingly you can play around and tinker with the volumes, but in general if you need root access then open shift will disallow it as a part of its security feature. You can also make a custom starter project like for example if you want to make a specific type of project template that belongs to organization that OK.You have this, uh, you know. Do you have this, this and this configuration? Your SRC file belongs here. You have a set of three other microservices depends on, so you can abstract this away from the developer by using a concept called is source to image still source to images? Nothing but you need to give a few executable files. For example, assemble, run and build in an STI folder within a GitHub.And then you need to give it a dockerfile.And you need to then extend an existing. You can either extend from one of the existing open open shiftstarterprojectslike.net node JS or Python And you can.If you want to.Go ahead an customize what is there in the assemble and build directory and then what should be running in the run directory. For example in the run directory you want to spin up a Dynamo DB database and you want to, you know link some micro service with the other or do some token exchange or handshake protocol with something else. Then you have to get into the world of Esther. Why Nesta Wise generally should be written by the operations team who have good knowledge in Dhaka?But many times people run into issues, for example out of memory in the bills. So your builds you need to go and you can go to the build and click on edit an you edit the resource limits. For example you can by default I think it just gives few mil equals an few MB of ram to build auto deploy you can increase that an you can also click on you know go to the building, edit the Yammer of the build so you can get into the fine grained communities.Controls with the help of the Yammer file an if you want to also modifier running pod you can actually go to the pod or you go to the deployment and click on edit deployment. Annual have direct access to the pod, the communities file and there you can tinker with things if need be, but in general you can. You know what your typical.You know, start up projects are in your organization or how much space would you need for your project so it will be good for you know to tell the operations team that OK, my Cuban it is. Could you please set this as a default while creating a cluster? And yeah, just be mindful that OK it is still using an easy to machine in the background or GCP compute engine or sorry or Google compute engine an GC engine so it will in current cost. So try your best to.Always think of Skilling so you can do OC scale an you can scale or an application to two pods or three pods horizontal scaling as their vertical scaling is also possible.By you know, increasing the number of CPUs that the particular pod pod needs.You can also set resource limits Ann. You can manage the underlying community secrets with the help of OC secrets and the overseas the overseas command line is very similar to cube SL. If you have work with Cuban at ease you can still do OC, get pods or see get SV CS or she gets services or see get deployments and you can still play around with the.Uh, the same concepts as community is. If you want to start getting into the nitty gritty’s in Python or to get a large model. Most people have this habit of using Git Alofs which is get large file system I. I think I had. I had difficulty with git LFS be’cause it did not work for me unless I made my own custom dockerfile and S2. I image. But those people who didn’t want to undergo the pain of making an STI image or just making an image pushing it to.You know the private registry of your project. What you can do is you can take care of some complexity at the application layer itself. For example, when the app starts up. Then you can use the bottle library to pull in from an S3 location, and then you can define one health check endpoint or readiness endpoint very similar to Cuban at ease and you will say that your pod is not ready if the model has not downloaded. So you can make an end point to check if.You know the file system has any file or not and that we large model issue could be tackled an similarly. For example, NPX node shift is one simple way to deploy angle application, but you can also use the node JS Express static framework to deploy the angle application so you can abstract away most of the complexity in the code if you don’t want to get into a dockerfile and the Cuban 80s part an just like you see TL you can exactly into the computer with OCR research.And if needed, you can tinker with the community side of things for the courses courses, I recommend that you subscribe to shape blocks from Openshift. He’s done numerous projects. An redhead itself has released numerous courses in Udemy there is one free course which I came across that you know makes you undergo the basics of Docker and communities and then it explains you what how you can deploy a simple application and open shift.In the Udemy course so have fun with openshift. I think it’s a very good way to apps you don’t need to know docker. You don’t need to know communities. You can still use the goodness of the same platform without actually having the knowledge. So play around with it. And yeah, get kickstarted. Be cloud agnostic. Don’t be tide to a single cloud vendor an don’t have to upscale yourself so much that you know all these concepts but you can still use them so.Reduce, take care.Thank you for listening to me by. I’m current bhandari working as a technical lead and associated generally.This is technology icing podcast goodbye.

 

Startup India demand for solutions: https://www.startupindia.gov.in/content/sih/en/ams-application/challenge.html?applicationId=5e79126ee4b055bfaea9ef66

Open source corona Datasets:
https://innovate.mygov.in/covid19/#tab6

Participate http://hackacause.in/

Some software engineering solutions during the Pandemic https://www.eu-startups.com/2020/05/10-innovative-coronavirus-solutions-created-during-online-hackathons/

Opensource genomic health realted datasets
a. https://cloud.google.com/life-sciences/docs/resources/public-datasets
b. https://aws.amazon.com/health/genomics/

Robo kit to fight corona
https://www.entrepreneur.com/article/349475

IOT Hubs compared
https://dzone.com/articles/10-cloud-platforms-for-internet-of-things-iot

Examples of IOT in health care
https://econsultancy.com/internet-of-things-healthcare/

Example of Arduino sensors
https://robu.in/product/arduino-37pcs-sensor-kit-set/?gclid=Cj0KCQjwzN71BRCOARIsAF8pjfivLYiYDsGYnMgQSnZCFBR8o0VmrjjsNCHE4ij2cDOPibGpQDo-pe0aAqIpEALw_wcB

Corona innovation challenges: https://theconversation.com/india-how-coronavirus-sparked-a-wave-of-innovation-135715

Make your own zoom/skype like platform for your org with the aid of AWS chime https://aws.amazon.com/chime/?chime-blog-posts.sort-by=item.additionalFields.createdDate&chime-blog-posts.sort-order=desc

Approximate Transcript of the podcast

Greetings and welcome you are listening to the technology icing presentation by Karen Bhandari and in this episode we will be covering how you, as a software engineer, how you as a system engineer, how you as a full stack developer, help in the Corona crisis.Well, most of your wondering, you’re working from home and you’re thinking how can I contribute? And many of you have this block. OK, I’m not a medical expert. I do know how to fix the symptoms. I don’t know how to. It’s the health line. OK, it’s not for me, no. I think this crisis demands help from across the gamut and software engineering. Such a field that it can be plugged into any domain.Any which way an everything interfaces with the software. If you are working on even a semi automated solution, it all interfaces with the software. OK, let’s cut to the solution ING. What are the problems?Our existing that you could pitch in you could contribute with your help. I know that all of you have unlimited resources at your end with the help of I’m not talking in terms of monetary, but you know computing resources.I application of services or systems or services you have digital ocean as you are a WS Google cloud platform all at your disposal for just a few cents. You could like spin up clusters and do things so.You could make your own solution running for almost free. Of course you can listen to my episode on how to use the cloud for almost free, so first solution.You could think of is a fake news detection. I think most of the people right now are flooded with recipes or tricks on how to overcome the virus. I think it will be good if you know they can get a faith score against it in the sense how verified it is OK. This has been verified by Reuters. This has been tried verified by Bloomberg where verified by.Or you know, some particle that Google News has got into. Or maybe, uh, how genuine it is. Has it been made by a deep fake we technology or not I think fake news detection is definitely something that the government wants to control. Not just in India but all across the world. Because it could be misleading. An it could strike panic. The second thing is logistics.Today there’s a lot of emphasis on no contact delivery of essential items of medicines, so there are products out there that assist you to do it. But so many enterprises want to get into this contact list delivery way. So what you could do is you could hook into the APIs of one of the existing startups that offer logistics, like for example in India there is done so there is.V is something that a career agency that starts with V care that is my portal for truck services and you could connect it to you. Know some Kirana stores to help them with doing this kind of logistics and also maybe add.A new twist to it by, you know, asking some NGOs to pitch in with, uh, you know, a sales of masks, sanitizers, an extra layer to just, you know, sanitize their equipment. The third one is cloud Management Today you should be able to figure out.How to manage crowds if if large people are not gathering at one particular point?Maybe you know sometimes you could also use a drone with heat sensing cameras. This is as per the startup India government. They have government. They have launched startup in their challenges. Take a look at the website and see what kind of solutions they are asking for their movement tracking to see how to track people. So if you would have heard of.I don’t guess you to have that was launched by India. I know it was a privacy nightmare. There’s a debate going on, but I thought it was very Noble and very time saving and innovative of them to, you know, think of such a solution because it’s very difficult to track who met home and I think are against it too, is a good way for.The government to find out how and who came in contact with whom. I think you could.You know also ensure that since you’re enterprises, may have could have applications that could help you with such a thing, but Beware of exposing publicly identifiable information because I think I would say to one French hacker had shown how it could access health data for other people, which I don’t think is advisory is, so just be aware of the security aspect.Geofencing is one way that could help you to, you know, track who entered which area or think Android and Apple have good APIs to help you with Geo fencing. Qualcomm’s also got good instincts that can assist you.The other one is called Comestic would also help you with finding out what are the hobbies. Approximate age range of the person, what kind of inconvenience they have. It could help you with those kind of information and then people want contactless entry. You know now you should not have fingerprint thing. You should maybe scan the iris of the person and allow him in.And or maybe read the Heat signature of the person, facial recognition for AI based contactless entry. And you could also help with.I show all the manufacturers right now they use a technology called C&C to manufacture items so CNC requires some some a computer program GM codes to be fed into the system, and then it makes.You know it makes the product higher by using subtractive manufacturing. Now there is another kind of manufacturing called additive manufacturing, so where you can use it free software like blender 3D an you know, make things that you could actually print. You could either rent the printer out in an agency for $20.00 an hour, $5 an hour, depending on which country or staying in there are many 3D printing windows out there where you can upload your design and they will print it out. Or if you could manage.You know a 3D printer for something like $70,000, and then you could also help with.These protective mask will you know, or maybe some equipment that can help to absorb the droplets. Some low-cost masks. I think you would have seen the plastic we’ll heading from straight from the head to the bottom. I think not to not take stream button but in the chest level. I think you could print these out because it’s small enough an.But then you know there are, uh.All these cloud providers like Amazon and Google have these genomic datasets that are published so you can see.Clusters and patterns of DNA sequences. That’s maybe an advanced level, but maybe you could go to a more.You know rudimentary way. Like for example, you could use Apple’s like like one company used Apple Watch technology to find out whether or not he would develop a heart attack, whether or not he’ll develop respiratory problems, whether or not they develop depression, because by looking at your pulse you know you have alot of fitness equipments. Today Fitbit Alta you have.You have some adreno base to Bell Street monitoring devices that could you know, like you could just have a wristband type of thing that goes around your arm and it could measure the pulse. An today mental stress is a big issue so.You could assist.With by using existing datasets you could train the animal models. You could also come out with so you know a lot of covert related datasets of also being exposed. For example, there is over 19 open research data set which has more than 30,000 articles on text mining so.So you could use that to make a chat bot four FA, QS. Uhm, you could make.I automatic health advising systems like if you know today doctors are flooded with messages because they have been urged in many countries to be available.By text messaging Becausr Today you can’t get an appointment at the hospital, so they will suggest you something that you could take from home or buy from the contactless medical stores around you to help you to fix whatever is going wrong in your body. And then John Hopkins University has also released there.Epidermal logic or data set and although European center of Disease Control has also released their COVID-19 data set, an also the hospital capacity estimation.Oh has also release the data sets, so you could find out if where is the nearest hospital for users for your company employees if they feel symptoms and maybe even take a self assessment test before you know actually approaching the doctor to rule out if it’s not and covert symptom an you know the World Bank indicators that are relevant to cover 19 is also available. I’ll put all of these in the.Show notes, then the Ministry of Health of many nations have released some data sets so you could see a lot of.Initiatives going on in that realm.You have hacker cause of hackers dot IN which has some competitions exposed there and people tweeting about solutions. So one very heavily demanded product is something like a ventilator respirator. So softwares around it so fast to interface with it.Or if you could 3D print some small subsections of it to help the company and you know supply to it that could be of help an.You could also see if, uh, you know if if you could give aways help the stores around you to go online to be available on, you know these E Commerce platform or maybe use some WordPress plugin to help them to come onto the E Commerce platform.Shopify experiences for them so that you know the UM.Oh, don’t run out of business and don’t run out of the monetary means to be alive. And many companies are not able to survive the lock down.You could also interface with the, you know some iot sensors that so as you’re a WS Cisco sales force.Have all have excellent iot platform. So how do these iot platforms help? Your device can initially register. So if suppose you have insulin censored or if you are also available. If you have you know, temperature sensor humidity is and so you can connect you to add Reno or Raspberry Pi kit an you could you know send or register that particular kit with one of these cloud platforms in a secure manner and then you do. You can also you can have.Aggregated view of the datasets that you have by looking at these iot platforms, helpers, enablers.And, uh, you could you know interface with this and make solution so you think? How do you production is it? I think if you just so if you go to Amazon, you’ll see Raspberry Pi cases. You’ll find adrenal cases and I think some of them have even got enough room for a putting a sensor in that. Could you know assist you with various things an you know some people have metal touch sensor distance in so.Temperature sensor infrared sensors. These can be interfaced with Adrino Pie. The joysticks for controlling drones. Now drone delivery will also be something you could also test if you know some people have also made some iot devices that you can ingest into your body and you know it. You can also see whether your blood is coagulate Ng or not and whether.You know your respiratory system is intact or not, which I think of Corona and the final stage causes you a lot of respiratory issues and there is a whole set of robots that you can get from the Robo App Store. An right small snippets are program to maybe just press if you detect a hand below, press some sanitizer to.You know, clean the hands, hand over a tissue paper to them.Today, most of the faith in a shop will be governed by.Are you able to maintain help the users or the customers to maintain social distance or not? So imagine a grocery store if they enforce 22 meter distance between each one an if the sanitized you before you enter. If you allow only 20 people in the store then then you know customer develops faith in you. That OK, I will not fall sick if I.6070% of cases. If I enter your store, be cause of the precautionary measures you’re taking an airline company who will just offer the entire suit to wear top to bottom, or maybe provide some UV cleaning and you know, be very finicky about health and hygiene. I think these are the kind of organizations that that help that you know may give the perception to the customer that you are being safe. I.Tink will attract customers like for example in Denmark when restaurant meat like an entire shell made up of plastic so that you know no other person could talk to them. Or you know the infection that aren’t could not spread to them. So these experiences that you provide will be remembered by people an.Banking I. In general, many of them have started exposing, you know open banking API there is PayPal that is tribe that helps you to exchange payments for contactless or cash lash kind of transactions an.I think you should. There are not just these.Solutions that need manufacturing even if you need to do manufacturing, it’s easier now with thanks to the help of.3D printing but even non manufacturing. There a lot of demands and there’s so many datasets that are there that can help you to. You know, detect clusters of types of people who could contract virus who could not contract virus maybe did you know do cross correlation with that? I’m like OK 10 days ago with this with this temperature with this humidity with this blood pressure this person was.Oh, in the 10th day of Corona, before showing any signs of symptoms, because many of you would have heard of 14 day or period before the symptoms starts showing. So you know you could take a time series data sets of various test results and come out with prediction models. You could use Google Colab too. You know. Test out your email solutions in terms of flow. It’s free of cost.Um?And Oh yeah, you know, just get hacking and try to help your organizations. Or maybe you yourself can close your nose and jumping into, you, know, providing solutions to help.The crisis overcome the crisis you have. You have a lot of sensors that you can purchase. If you want to get into, you know some more.Vis, in which you could do.Contribute an even in messaging solutions. You could help with summarization using an LP techniques for some people who are quite busy, like for example doctors.Uh, what transcripts and a lot of things could be summarized with the help of an LP. So get cracking an don’t be, you know, distributed by the fact that you are not a medical expert. I think everyone and anyone can contribute.To overcome the pandemic, I think I have recorded this quite late we.I have already crossed 40 days into the pandemic and I think it’s never too late be cause even post this. I don’t know. Whenever it ends there will be a sense of caution an we don’t want you know, a relapse of it. So get cracking. Hope you come out for the innovative solutions. Take care, bye, stay safe, stay at home as much as possible.

 

 

Approximate Transcript

Hello everyone and welcome to episode 16 of the technology icing podcast you were listening to. Current bhandari or technical lead in so sheated generally and in this episode I will be covering office Productivity Tools.The most important office productivity tool I feel, is GitHub.You can go to the project section and.Have a Kanban try type of food where you can have leans to do in progress done.Very similar to Trello, GitHub can not only keep.The source code.But also manage your projects just like how Trello is doing.GitHub I’m also using it as a documentation engin. There, Vicki Section. You can make markdown files, and if you don’t know how to write markdown files, you can always click on create new issue an. I’m using GitHub like an image server, I just drag and drop an image that gives me a unique URL in my personal life I have this unlimited storage given by GitHub an I give whoever I want, I just give them the URLs I drag and drop any image to the new issue. When you click on issues new issue.An I can compose text, bold it underlying it, have headings an insert links. You know I can just drag and drop a file like PDF or document and or maybe a zip file an I can share it with someone who cares about the limits that are given by Google drive or one drive. I just drag and drop stuff that I can make links. There is one important file format called markdown. I even write my emails on GitHub.And then I copy paste on outlook or Gmail and that way the other person receiving the email gets a lesser size becausr all your images files are hosted on GitHub.Please note that your your URLs and images will be public. If someone gets to know your URL, but if your URL is not known to the public, no one will get to know. Very similar to Google drive link sharing.GitHub has many more capabilities. For example cross team collaboration.Even you know send private messages for Krepo’s work on open source solutions, not just for developers, but I’m using it for documentation for wiki pages as well as project management like quick agile Kanban board can be made in GitHub Project section.And yeah, if you are looking for a more dedicated tool, you have jira with you. Angira has API provided so you can, you know, send an email as soon as issue is created or maybe started build gira, gira and GitHub could be integrated with each other. You can mention task you can work in an agile fashion, define, you know, get lot of charts like team velocity an track your time there, a lot of organizations are using data.Track people in the sense today, if you have eight hours we created to has called meetings, you create a task called lunch and you create a task. You create individual task for the development activities for office related activities. Even for labor work. You people can go into the task and click on. OK I want to log to us today. I want to log one out today or half an hour today. So that way you’ll get to know what the person or your Co developer or your team members doing within the 8’s.And if you are you looking for a free version like a free premium version of jira vests, Microsoft Teams is extremely good. You can not just maintain you know your project documentation that you can also integrate with GitHub as well. As you know, start builds. It gives you a free build server for up to five users and you can create task in a very hierarchical fashion for free. I mean, Gina is a paid version.But I take Jira cost around 10 users 10 USD but vests. You can start for free teams or devboardasyour.com. It has different names we used seems an you can have a hierarchy of epics under that features and stories and issues and tasks. You can create this kind of hierarchy and even Wii Kay Reich is quite good with team management.It gives you 2GB of storage with Google drop box and has some basic task view WRIK and there are some open source task management solutions to if if you want to use Trello. I think Trello offers you a free only Kanban only view, so you can only view one board where you can move things from to do in progress.Done things like that, but then Jira is. If you want. Jerrod vests can give you that hierarchy that time management view the project management view and I apart from the office tools like Jira and GitHub, you you will spend a lot of time in emails.And emails today the only thing that I see in the offices I have about 3000 unread emails. But I try my best to get to the more focused email so outlook has released a new feature called focused and non focused. So it will try to take out, you know system generated emails like for example Jenkins is generating emails you get you know newsletters. So it very similar to Gmail’s promotion stabs outlook has this feature.Annum I feel that if you um don’t want our specific service you just want to sign up for a service but you don’t want to check the emails of that particular agency when once again you can use a service like mailinator you can create an email like your first name dot last name at melanated or man it’s a public thing OK so make sure that if you want to subscribe get promo codes get coupons use mailinator don’t use your official.Inbox, but then this is public in nature, so it’s OK for the marketing emails to to be public. And if you get an OTP.Uh, make sure it’s not connected with money. It’s only connected with activation of your account. Make sure there is no account that you created mailinator that is associated with credit card. and I think you can use more.Uhm, you know better you know more organizing tools that sits on top of your email. For example, get organizer. It does not have support for Gmail, but any M PS provider and get organizer can clean your inbox and categorize it to different folders. There’s also something called a screen email which is $2 per month. That can help you to organize automatically based on some keywords. And of course outlook has a rule based thing, but in using other services you can use some milk based approach to.Learn what you are doing and then at remembers by itself and while composing email I told you just composing the issue page of GitHub and paste it in the email your formatting is taking care of your your you have more control of. You know things like spacing because Have you ever struggled with copy pasting from something and pasting an email does not fit very well. Then there is Grammarly.And Gramma Lee helps you to, you know structure. Your sentence is and make sure you’re not using. You know some less powerful words, for example.It will create more impact if you if you use some alternate words that are more. There are slightly more active in nature rather than passive. For example, I’m handling this issue.Is a slightly less powerful compared to.You know this issue is being looked into an.I think you should try to make your use an active phrase rather than a passive phase so that it shows that you’re in control of things. An Grammarly may be able to help you with such things.I know in office um Ms office is the best, but then if you don’t want to pay for the license then Libre office is there but it’s not there yet. It’s you know a lot of formats are not being transferred over. I think Ms office is a necessary tool that you have to get. You can write Plugins, for example Excel plugin using Excel DNA and you can write formulas by yourself using C Sharp code. You can use macros an you can use C Sharp to.You write Plugins, JavaScript to write Plugins in Ms Office. Ms Office is damn powerful. Of course. Google is a very important content of they have APIs as well. If you’re using the Google Business sweet and so who Mail what I do is when I purchase a domain I if I want FREE Email address is 0 is so who gives 5 three email addresses?I know, yeah, no beyond office your your email.There is this. WhatsApp management?What one could do to hand?Will WhatsApp is OK connected to whatsappwebweb.whatsapp.com to connect it your desktop and that allows for easy copy pasting or you know replying to multiple people at a very rapid pace. However, if you want other people also to manage your words have for example call your wife to manage with you called one of your secretary to manage WhatsApp because I’m receiving 1000 plus messages then you can use.Call Bell CALLBL or KALEYRA Kyle era.And of course with WhatsApp, what next comes to your mind is photo is Adobe. Adobe Lightroom is a paid version, or you could use Picasa. Or if you are feeling slightly more powerful, like if you feel that you know you could use algorithms like machine learning algorithms to organize your photos, you can use Microsoft Cognitive Services an they are available for per photo. I think they will charge you sent.And no, they can automatically organize all the photos for you because it can give you the key words if you just upload an image, it will give you the key words and then based on the keywords you can sort it to directories. There is jet photo studio which is free.And the next in line to photos is files. You can use CC cleaner to remove your duplicate MP threes documents. There is all dupe CC cleaner can clean your web browsing history, temporary registry, temporary files and just you know after every Microsoft upgrade you just all of a sudden windows just talk starts occupying fight and GB here and there. I think it can help you. Then the other one is accounting.I’m not very well aware of the.Speed versions of course tally is quite popular in India, but but yeah, for the free ones I use Jianu cash. There is zip books or wave or OK credit dot IN web based one and here you can maintain different accounts for example expense, income. You can group your expenses into various sub categories. You can have different kinds of income accounts maintained and then it can generate balance sheets in the end or profit loss statement.And next is events. Uhm, it’s easy to create events using town script for event brite and you can easily integrate with Google calendar and down script in Eventbrite can help you to even have an organizer app and that organizer app can scan a queuer code because every time someone joins the event you can have a free event or a paid event and if you have a free event also you can use it for attendance tracking so it’s very easy to.And you know all event related messages could be put into event Brite.And now this ends event section. Next is, you know, just team collaboration vests as I had mentioned, is the best. The second is share point which has a jive SharePoint OneDrive. You can create your own and Office 365 course you can create.Your in house YouTube. Your in-house document management system in house idea management system. Like you know, create a topic. People put ideas, create discussion boards create. You can just keep your files in it. Keep videos in it, group it by topic by teams, by projects within a team. It’s quite powerful using either Jive, SharePoint or vests freed SharePoint OneDrive appeared.And if you want to store a lot of artifacts, like for example OK, you made something you want others to use it soon are Nexus is quite good. It’s not just you made something and someone can use it just like that. But you can also manage DLL’s Jr files, have your own maven repository, have your own MPM net worth a tree or new git repository. And so now if you want to keep it for yourself like a private version the Sonatype Nexus.Is available as a docker image.Sometimes you just want to run a piece of code you don’t want to install. You know Visual Studio VS code via Square is a most powerful tool. I use it for markdown for GitHub as well. As you know, all kinds of programming languages java.net Python VS code is enough. For me, Visual Studio code. I think that is the best tool. But then if you don’t want to install any tool and you just want to run, you know.A piece of C Sharp code piece of pipe.If you wanna just run code then you can use a WS Lambda or as your function. I think up to 1,000,000 times you can run for free and after that they can. They’ll charge and sometimes you just need a machine for five minutes for 10 minutes for running some job or batch. Or just run the computation without you know you’re running something. I think a WS code build is an amazing thing, it just gives you a machine for two minutes. It takes the script file like a shell script.You could make the shell script call or piece of code an without you installing any software you can use a WS code bill for one minute 2 minutes 10 minutes and it will charge you just one cent.No per minute, which is very low. And of course if you want to use development environments, docker with elastic unity service and as your community service is amazing and just in case if you have too much free server space and if you unfree, computing resources becausr your employees sleep for eight hours in the day, you can host it.And now you make money out of it. You cannot make it available for people to host websites. There is WHMCS solutions, we have host management control systems that can help you to monetize on these assets. These are some of the office productivity tool kits. I’m sure there are many more that people are using for their day-to-day work.We’ll see is damn good to reduce video size is.Image editing paint jimp is good. Inkscape is good if you want it to be. Uhm, you know, zoomable, an be like SVG vector format.Or did I many more tools that are there in the office Arsenal, but in the interest of time I am going to end this podcast now. Hope you enjoyed it. Take care and wish you a productive era ahead. Take care bye.

 

Important links
https://dev.to/sheddy_nathan/possible-ways-to-reduce-your-webpack-bundle-size-js-secrets-550

https://survivejs.com/webpack/what-is-webpack/

https://blog.jakoblind.no/3-ways-to-reduce-webpack-bundle-size/

https://webpack.js.org/guides/code-splitting/

https://blog.jakoblind.no/webpack-code-splitting-libraries-what-to-do-with-the-files/

https://itnext.io/react-router-and-webpack-v4-code-splitting-using-splitchunksplugin-f0a48f110312

https://medium.com/react-weekly/code-chunking-with-webpack-a-pragmatic-approach-e17e8bcc6453

https://jeremygayed.com/dynamic-vendor-bundling-in-webpack-528993e48aab

Approximate transcript
This is a technology icing presentation and you are listening to Karan Bhandaru and in this episode we will be talking about how do you reduce the size of your web application – more specifically reducing the size of your web application in production.When I say web application, I mean, it could be.Any angular or react application that you are bundling with the help of something like webpack.Most of them went back an hour that is reacting angular provide you something called as production mode.This production mode is something that.You could use so for example, if you do webpack minus B mode is equal to production, it enables something called, is uglification it enables.It you know magnifies your file and it also you know make sure that your long variable names are converted to slightly shorter variable names.And typically most of the fatness in application comes from Lou Dash.An in some cases comes from moment JS because you need to maintain international isation of with various.Products, an moment, has very good date related functionality. So I would recommend that you use a slimmer version of low dash for example, there is low dash slash get low dash slash concat so you know you don’t have to try your best not to use lodash. But if you still want to use it use the minified version and you can also install something called a slow dash webpack plugin.Um.And there is something uh and and when you use it. You can you know put it in your webpack config JS that will do that work for you?An try your best to use ecmascript 6 module centers in tax for example, import connect from react redux instead of you know, some doing something like import everything from react redux like just import the entire library don’t do that just import a specific part from that.And then you know webpack will ensure that only those parts are included in your final output.You could also check what is causing your web application to inflate so much there are very good tools like Webpack Bundela Analyzer?And source map explorersweb backbend bundle analyzer shows a graph and a heat map of what is causing what to?Expand sometimes what happens is if you include high charts and that specific module of world of world map can be extremely massive so it will be good if you can you know split the libraries as a separate bundle?Or maybe you know fetch the bundles asynchronously. I’ll talk about that, more in the upcoming minutes.And No.In order to split there is something called as webpack junk manifest.Ann Webb back there is something called, is a manifest plug-in.An it will include the file names of a generated bundles, including the hash and then you can refer to your to your chunks in the script file like which kind of chunk. You want and the other alternative is using something called?Good there is something called as you know a waiting on the import so typically most people just put imports on the Top, but there is an article by Webpack and that talks about using import from your from within your function imagine calling await import.Your Lodash then what will happen is you’re importing of that specific module of Lodash happens only when the person views the page.Um.So you can you can load some things this technique is called? Is lazy loading?An this is quite helpful in reducing your bundle size. So Suppose a four person does not open. The particular page where the world map is required. Then it may not be loaded only when the specific pages loaded then it has to be loaded.And then there is something called as you know.Split chunk plug in and the other one is.There’s something so here you mention.Here you mention what?Chunks you want to split your so it’s an optimization dot split chunks and you mention that OK. I want everything in the node underscore module folder to be as a part of vendor. GS everything under the so slash Maine to be a part of the main application, so slash subsidy to part of the subsidy application.And then you can use that?Of you know import syntax to await on the import from within the function to get that particular file. It’s called dynamic importing.Ann Webb back has got very good or bond analysis tools like Webpack Mundella or optimize overpack stats plugin.It’s quite intuitive to run it.Ano split chunks plugin is what you could do an?Make sure that you use the you know the smaller version of lodash not that.Like you do low dash slash thing instead of doing the entire lodash. It’s it’s.It’s quite helpful and there is also something called us common chunks plugin.Um an it’s it’s given by the author of a pack and it accepts a callback.An you can also include that in in it for and there is something always differ. So you can differ your some of your JavaScript files that are required for so suppose if you have done a chunk of world map. Then you, you do not load it. At the time in the script bundle you can load it. Sometime later and you can use that default plugin and there is there is a difference between a sink and Defoe in your scripts.Scripts were you mentioned the source you can put before there an that we the amount of fatness in your JavaScript could reduce.So a tree shaking is automatically enabled in Webpack, an that is only possible if you use?This one sometimes you may be troubled by CSS S so you could bundle your CSS S into a separate file an you some other optimizations. For example, if you use SSA SAS less sorry sass or less that way you’ll be able to know what is the common functionality?And only that particular functionality so you know you can think object oriented in your CSS S also and that we will be able to.Reduce UCS S files by knowing what to actually.Import and then you can also look into advanced uglification options because sometimes if you uglify everything. Some PDF library example. For example, is not working with uglification then you can you know go to uglification plug in and?Tell them OK, I don’t want to uglify this particular one.Ann.Sometimes there is also something called, is react code, splitting if you want to, if you want to go ahead and Dooley could react specific code splitting so look. It look into the Webpack Docs. There is also a very good book, I have mentioned in showing words.Um.That talks about how to?Go about using webpack in a more advanced fashion and there’s an excellent site called survive GS survive GS as.Extremely good tutorials on how to use webpack. I think as in advance user when you start running into issues regarding you know side being too slow and what you also could do is sometimes you are as your may be hosted in US eastern in your users are in Asia try your best to move this vendor files. These the SRC files into a CDN.So that way it is kept in a location, which is closest to the user and the network need not made the round trips all the way till?US to get that particular JavaScript file.And try to use HTTP 2, features by making sure that your JavaScript files are all under 1 MB.An don’t use things like.Bootstrap use bootstrap min dot men don’t use.The normal libraries that are available for debugging use the ones that are the minified versions, so an also think about your images images could be made into something.Images can be combined into one image and use like a Sprite Amit CSS S has got something called a CSS spriting.So if imagine you have Alpha imagine you have a character with different moods. You can put all this character into one image like small small for Mickey Mouse is and then you can only render a portion of that image so that way instead of the.Website loading 100 images, it could just load that single image.And make sure you use?Proper image optimization for example.It’s it’s OK to render smaller images on the phones and larger images on the desktop.You could use.10s of floor to you know crop transform rotate or Imagemagick to.Process your images in bulk.And there are some impressive image handling tools.That are available.I know try to use SV GS is instead of images as much as possible use tools like Inkscape to create an image and don’t use photo shop because if you use SV GS then even if you zoom in by 200% or 300% it still look crystal clear.Um.And uh.Sometimes it’s OK to run your application.Are some of the compute intensive task? For example, you’re running a for loop on 2000 numbers? Maybe transfer it to a web worker.Ann.If you have a lot of data to be stored go don’t store it in the rammed as such.Keep it and index database in HTML5.Ann.Maybe use a little bit of the local storage.Try not to keep everything in the ram because the UI thread is quite valuable.Ann.Sometimes you can use signal have to.Meet the application look very responsible like for example, you need not pull. The server and ask are you done are you done? Are you done like for something which is very long running you could just ask the server to send you a message.Make an and make sure that you implement throttling in your website so that some irresponsible users don’t.Floody website with too many request that way your server is also responsive to you.I know it’s good if you wanna use start using Geo replication.For your data in the server and try to make your UI fast.Sometimes it’s OK to store the UI elements.In S3 or blob storage.Best is to use a CDN so Verizon CDN, Google, Microsoft have extremely good presence across the world.So keep cracking yourHTML5 web application size visually pleasing coding session, and hopefully a users never get annoyed because of slow application implement the steps. An yeah, happy coding goodbye. This is me. Karan Bhandari, signing off. I work at associated with Socgen as a full stack lead developer goodbye.

 

https://github.com/weaveworks/eksctl

eksctl create cluster --config-file=file.yml

Approximate Transcript

This is the technology icing part cash powered by current Bhandari and in this episode. I’m going to cover the topic of Kubernetes in AWS  which is generally driven through E KS elastic community service.No let’s try to explain or typical scenario for example, or retail bank wants to open.A set of services available for the customers to consume for the mobile apps for the web block applications. You have a simple transfer screen annual have a simple.Payments module so let’s assume you’re building. 2 Microservices, an your payments module has something interesting. It accepts our check scanning an you could deposit to your account also apart from person to person pay so you would typically make it in this fashion, you would have.Are transactions micro service or payments microservice ,an accounts microservice an image recognition as well as OCR micro service?So, in order to go about.To start composing everything you would know have to know about a concept called’s Cuban it is deployments.Deployments is something that.You mention that I want 3 replicas of account service becausr accounts would be used not just for the transaction, but also in the from and two part of payments. You could say that you want 1 microservice of one replica of Microservice.Of accounts.And since you feel that.Your image recognition shall you solution would take more.Our resources you could also define how many nodes.It could occupy.Tensor flow Python has support for coordinator which can allow the tensors to be flowing through multiple nodes.And if it’s done poorly.I know in order to go about to do that. You need to define communities eml file.You would first need to install EKS CLI.Which is available it’s called E KS ctl eksctl ?And you need to ensure that.You have installed the latest version of Kubectl is available by doing a curl command and then you can also install mini kubectl so that you can do some testing in the local.An mini kubectl or depends on something like VM Ware. I have not seen much success in using the bare metal OS.It’s easier to do all this from a Linux machine. I feel it’s better to partition your machine.Into a Linux and Windows 1.And or maybe just work from easy to low powered machine to run your capabilities workloads.Of for only thepart where the kubectl see tail is installed, not where the actual workload is running.An certain instructions you need to keep in mind are kubectls. ETL get pods or kubectl. Cuttle get pods. kubectl get pods or kubectl get services kubectls get secrets.So these are some things that you would keep in mind.Oh.For finding out what’s running in your kubectl  environment because kubectl see tail is the command line interface that talks to your pods your clusters your services your deployments you replicas.Ann.If you ever need to debug you need you may need to, you know find out some more advanced commands like you see TL describe pod.Or kubectl you can also do log so you can even do kubectl log.I know.You could mention the pod name like you, said your logs the nginxX or whatever the pod name is.Ann.For troubleshooting you may also need to know certain things like how to exact into a specific machine like you see TL exact and then interactive terminal it and then the pod name and then you mention you want to Exec as a bash or a shell script. You need to have some degree of comfort with.Linux command line over here.An yeah, just be in general aware of how to handle things like for folder permission issue make sure you have the sufficient rights.And No.In general make sure that if you are getting a message Pacific to Amazon like failed scheduling. That means you’re giving it a very low powered machine like T2 or 3 three or T3 Micro. Maybe you need to go to a T2 Medium and if it’s coming as field scheduling in even T2 medium. ABM file large is something that you need to go OK? What are these these are nothing but?Amazon machines with M is memory series T so, so you know the ones that are provided your account for free or almost free during the first eleven months are the T2 Micro and T2 Nano series.It’s better to first start with making a docker compose file so that you can get a hang of how your docker containers are working.Annum you if you if you were OK, with publishing your.Artifacts or your docker images to.Public then Docker Registry or docker hub is a good option, but I feel easier is a good option an in ECR,I feel you must not load.Everything that the container wants or sometimes what happens is for example, Python needs or training file like an H5 file or a spark solution needs some HD FS file which are typically numerous GBs.I feel you must load it.During runtime as soon as the server starts as soon as flash servers running or as soon as Park starts running you can start fetching the file from an S3 bucket because keeping larger files in easier is too expensive. But if you loaded from S3. You can even you know control whether the traffic hits your particular part or not or your container or not.The traffic is something that can be controlled by something called’s liveliness probe and  readiness probe.And liveliness probe is when you APIs Radiant readiness is when you are OK for the Kubernetes Controller to start sending traffic to you so you can mention you know what could be your liveliness or readiness group. Typically, you could make an HTTP endpoint or some file system related probe to see if a particular file exist or not, and you can define that in your Kubectl deployment or the pod definition. so you could try making and ensuring that your easier images are slim so that even your building your deploying time is reduced. But it’s just that when the traffic starts routing it takes his own time.Your typical.So how is your services structured you would want to keep most of the things I was internal as possible, unless the application really wants it.Generally it’s difficult to you know expose every service So what I would recommend is if suppose you want to expose 3 services for example the account service you want to know you know you want to give the ability to see the list of accounts and transactions service you want to see the transactions.An image recognition you want to expose the point where it detects which paths which objects have been detected. Then you know you don’t go with adding exposing all your services. You keep all your services within the same cluster and then you maintain a new service called Ocelot, which is like an API gateway so in API gateway. You could define Ocelot service. An you could mention that servie to be load balancer with that ocelot definition, you can tell OK when I get my service and will you know, do like a 302 re direct to it, but the client won’t see it within the API gateway, Desiree direct gets the data. So now the best part about it, you just need to expose your API gateway.And all of those are clustered IP. So, your API gateway is something that has authentication. You could use Cognito or you can use Azure identity, or you could use auto an on just your API gateway an?Then the other others are are not exposed at all. The edges running in your cluster IP so even while during development. You don’t have to worry about authentication because all you are 3rd party. Traffic is going to be exposed over this API gateway. API gateway is available as a deployment. An you could write a service of type load balancing if you deploy it in.AWS ELB elastic load balancers already is by default provision to it.And dumb.There is there is an area, which I think Amazon providers own annotations for you know, specifying like what type of certificate would you like to use so every resource in?Amazon is controlled by.Something called as I am an?This I am can protect your Amazon resources from being utilized to what you could do is you can give that particular poured running and I am resource to pull the certificate or during the build time Amazon provides some annotations for the load balance.A foot example, you have an annotation, which is known as our service dot beta dot communities iOS slash AWS load balance. SSL Cert, you can provide that as an annotation and then you can provide them. The ARN resource of the certificate you have loaded in AWSAC and Amazon certificate manager.I know you could mention or you know the certificate. So, your your external load balancer.Which is of communities type load balancer?Everything else is a cluster IP down that load balance is the one that is exposed over HTTPS that could be a very simple solution. Of course, if you have you know a lot of URL rules?You could brightening English controller, but I recommend that you are you stick 2:00 o’clock doing most of your URL work.I know you could introduce other kinds of.The other services that have discussed his Cabinet is deployment. An images could be pulled from EC are and what they could be built is so you can have a WS code build to build to build you know your source code and converted and you know like do a docker build.And uh.Through the Docker compose and then push it to the East Amazon easier registry using code build.And then using code pipeline.And cold deployment you can deploy the solution using code deploying code deploy. You just have a series of shell scripts after install before install.And your code build could be written as a yamil file. Even the code deploy can be written as a yaml file an the code deploy and the code build so the code build will build it and code deploys the one that runs all the EPS commands or an EPS is very ESC. TL is a good tool for creating your clusters for example, I need 3 machines of T2 medium you can use the.EKCTL for it and for the for everything else. You would use the kubectl see tail commands for example, you would use kubectls. It’ll apply an pointed to your deployments directly, which has all the deployment eml files or kubectls details applies service.Annum.Are you could get the kubectl? To you know? Talk to?Your.Cuban 80s controller, so that it could orchestrate and run your workloads.I know.I would recommend that you would maintain your secrets in systems manager.Systems manager specific product called, is para meters 2, which is given by WS it’s it’s it’s free of cost.I know you need to maintain AKMS Key KMSKMS Key is paid for.But thenthe Parimeter store is free an but in order to encrypt values in the parimeter store you need to.You know go ahead an?Ensure thatOK, Ms Keys, present so the para meters. Talkin then while you’re running your kubectl create secrets. Command can be pulled from para meters to see your para meters to conversion your secrets. You can put their entire Jason value, but of course, it should be under 4 KB.Unless you wanna make an advance secret which is paid for so standard secrets are free.An make sure you know how to tear down your cluster using kubectls detail as well as ESC TL.Ann.Make sure that I you know you’re monitoring your nodes like by cloud watch on your application logs are being written to cabana.Ann.You could also he do an automatic code review by.Asking the code built to optionally run or sooner, scan that can show you, your code coverage in terms of test as well as how mature your code is.I know.Then there is something that’s to choose the code deploy could build an your.Code and your GitHub together what you could use is a code pipeline. Anna code pipeline allows you to even take approvals before you actually do the deployment code pipeline or allows you to you know set up a workflow that OK, when this change happens. Then you pull it into code build an you start building it and then you deploy it. But before deploying you may want an email approval, you may want to start.Some load balancing testing or your spec flow scenarios so.It’s good to you know use an ecosystem. You can use Jenkins for everything of what I would have just mentioned but I would like EW still manage most of it.A dub I feel that sometimes it’s OK for you to use.2 different cloud providers, but I recommend that you don’t do it because it will take time for one cloud provider. Talk to download cloud provider be cause of the different Geo locations. But sometimes so for example, if you want to use a cheaper solution like?For example, if you want to use the free part of Azure Web app. I think that basic app services free, which is 0.5 GB.And then there a AWS databases free the Dynamo DB so and initially if you want to start everything free and then you have your task written in dev.azure.com.So some people would want to start free and then move on to.Are you you know go to more costly resources?Then it’s OK for your dev environment to be like that. But for your you 80 and prod make sure that you.Don’t talk across clouds.Because it’s difficult for you to maintain slow or fast traffic. If you have a UI. I recommend that you put all your resources like the JavaScript CSS S in a CDN.Ann.Make sure you even within the internal IP internal cluster don’t talk using IP addresses use the name of the service. It’s easier.An I came across the situation where the services were not being able to discuss with each other like talk to each other typically or metadata and label is something that is very important to make sure that all of them have the same labels. Then you know the UAT environment can talk amongst each other. The dev environment can talk amongst each other. An yeah, the volumes I as fast possible try your best to use empty.Volume.But if you have a an admin with your DevOps engineer with you try to start the storage class PVS like provision volumes.But if you don’t have that luxury with you then you can just use the provisional volume and then you can plug in a cloud volume like.ELB or you could use as your file storage.I know.This way you will be able to maintain you know files and it’s it’s easier if your application is able to talk to S3 and pull it and put it into local either emptied our volume try not to use the host Derby cause that depends on the machine.But of course, if you just want to pull once.And not pull it again and again, then it’s OK to store in the machine.That’s it from my side, an if you have any doubts you could reach out to me on Twitter.I’m available at the rate of Ku RTZACE.KURTZACE that’s my handle, and my name is Karen Bhandari. I’m working as a technology lead in society general.Hope you have a good time, migrating to a community’s workload have issue productive rest of the week goodbye.

 

Show notes and Links

https://angularfirebase.com/lessons/sharing-data-between-angular-components-four-methods/

Parent to Child: Sharing Data via Input

Child to Parent: Sharing Data via ViewChild

Child to Parent: Sharing Data via Output() and EventEmitter

Unrelated Components: Sharing Data with a Service

—-

Reference passing

<app-side-bar-toggle [sideBar]=”sideBar”></app-side-bar-toggle>
<app-side-bar #sideBar></app-side-bar>

one object to have a publish/subscribe interface- windows opener example
https://stackoverflow.com/a/20526386/5059472

Thunk
https://daveceddia.com/what-is-a-thunk/

Selector https://medium.com/@matthew.holman/what-is-a-redux-selector-a517acee1fe8

https://www.freecodecamp.org/news/an-intro-to-redux-and-how-state-is-updated-in-a-redux-application-839c8334d1b1/
https://cdn-media-1.freecodecamp.org/images/1*VLQNO9Apn9qfm6BPYXG8TA.png

Effects
https://ngrx.io/guide/effects

SNS https://github.com/KMIntelligentSystems/SNS-web-worker-Angular-2

Ngrx socket io
https://github.com/avatsaev/angular-contacts-app-example

Signal r

https://github.com/Odonno/ngrx-signalr#readme

ngrx index db
https://www.npmjs.com/package/ngx-indexed-db

sync-your-state-in-local-storage-with-ngrx
https://medium.com/better-programming/sync-your-state-in-local-storage-with-ngrx-9d6ceba93fc0

https://angular.io/guide/service-worker-getting-started
A service worker is a script that your browser runs in the background, separate from a web page, opening the door to features that don’t need a web page or user interaction. … Before service worker, there was one other API
that gave users an offline experience on the web called AppCache.

https://medium.com/bb-tutorials-and-thoughts/communication-patterns-in-angular-9b0a829aa916
ng store
We can have a shared module which is imported in Modules and use the services to share data.
Communication using content projection

FILES
https://developer.mozilla.org/en-US/docs/Web/API/File/Using_files_from_web_applications
createObjectURL

https://www.html5rocks.com/en/tutorials/file/filesystem/
https://w3c.github.io/FileAPI/#slice-method-algo

 

Approximate Audio Transcript

Well hello Internet technology buffs. This is technology. I think podcasts brought to you by Karen Bhandari and in this episode. We are covering the topic of communication within angular applications.Our communication.Within an angular type of ecosystem like react Redux.I’m going to talk about the framework level. Things also because most of the production applications. The web applications are generally built on Vue JS or angular or react so I’m going to take the popular choice. Ann just talk about how do we communicate between components between servers between clients and just? How do you go about passing data?Well, in Anglo if you see the fire base article that I have given you in the show notes and links one can do parent to child communication by the input tag and where what you do is you just define some square braces in the.Tag or the element the custom element that you want to send data into an the other. One will have annotation on Top of a variable call at the rate input and they receive data.The other way is uh you may ask how will the child communicate back to the parent for that there is something that is a concept called, is event emitter.And for that you need to attach an output property an all you have to do is pass it into the output property dot emit an you pass the object into that emit function and the other parent will receive it. Outside using the non square braces, which is like the curved braces. So to pass data into a child. You use input to pass data from the child to the parent you use output?OK, this is for simple communication and.The more easier way is by just injecting a service into both of them.For example, if you just want to maintain the username or the user profile. You would just maintain a user service and simply use it as dot get dot store use angular properties. An you can use you can just use a global variable on a service an that’s the easiest thing to do.No, but I will also define something called, is a view child an in view child? What happens is you have reference to the.Do 2 to the child and you suppose if you are injecting something called as a child. A new element that you define within your HTML template. Then you can reference it by using something called as a view child annotation and then you can call anything of the child like the child dot something or child dot message child dot set something and you have access to the entire child elements.And uh.Yet another way of doing things like if you want a child to call the parents. Things then what you can do is in in the same input property that I had talked about in the first you pass on.You pass on you know a reference of something else like? How do you how do you reference an element in angular you just put the hash and suppose you have app sidebar you put hash sidebar and then the other one is supposed supposed navigation supposed navigation. Top horizontal line and you can just pass the sidebars reference into that and then one component can call the reference of some other component.The moose term logical thing to do is you know rely on a framework not use these bare bones things because very soon. You’ll get lossed, an which component is sharing information to another component an?It gets it gets very confusing so I would recommend you to use a frameworks like Angie RX where you define.You know a single story.A single store and.You know there is only one way communication only the reducers have capability to produce new store. You can’t just you know mutate the store. Just like that. You need to generate a new store and you don’t have to be terrified about you know, maintaining a single store. Becausr each each store can be subdivided within that OK. The store is supposed to be the single source of Truth, but logically you can.Separate them like you know that index dot JS file can refer to many elements so you can separate them by saying like OK. This is handling the stock ticker. This is handling the graph that shows the stock over 5 years. This is handling the news related to the company so you can maintain 56 toes and then reference all of them into the single store. Ann you use a framework to go about doing NGRX on Top of it?So NGRX is maintaining you know it should be a read only state an only if PR functions are supposed to update the state so this is in general. What redux is redux is nothing but you have a store, which has state an reducer is the only one that returns a new state and updates. The state and the action is the one that takes you to the radius and the reducer returns a new state so the UI.Can trigger actions like OK a button has been clicked?Nope, someone has opened a sidebar so you go and you go and start the UI keeps creating actions an you know reduces returns new state, but this paradigm is sometimes not enough. You you tend to have more complicated scenarios for example, on click click on often action. You want a server call to be made that just can’t be put into the reducer for that. You need to use something called’s effect.An actually before the effect came that unk so let’s talk about tongue and then I’ll talk about how.Effects can simplify tongue so tank is nothing but a function that returns a function. So suppose if I say get so suppose if I say that get score of a particular game. It won’t instantly. Get me the score. It will return the function that calls the HTTP call over the get in the sense.Like the trunk is a function that returns a function and this you can install a middleware and then your reducer and action can become smarter by not just updating you know by creating a new immutable state why with the help of immutability GS or with the help of things like the spread operator. You’re not just limited to that you can also use thunk as a middleware to.Make these calls, but the easier thing to do is use something called effects and effects is something that you use a method called, is create effect, an you pipe into the action an you tell OK, when I get the ad book action, I will.Go ahead and make a service call and then once it’s successful. I will fire another action then that action can update the reducer so an effect can be fired that listens to actions and then it files another action on success or failure.This is one way or this is a way where long running operations in the.US redux is possible won the simplest way is thunk, but no sorry the complicated waste hunk and the simpler ways effects. So NGRX has a guide effect has a guy to effects. I’m putting it in the show notes and.Most of the applications are riquier playing around with not just you know their data within themselves within you know communication. They also need to refer to server so for that. Of course, you can use a service rap it up with HTTP you can rap. The HTTP within a service. An yeah that service could be called from a chunk or from the effect.How however there are some more or scenarios for example, if suppose you have a fire and forget call or maybe a chat window. An you sometimes after 15 minutes or 20 minutes of server wants to talk to you, you just.Well hello Internet technology buffs. This is technology. I think podcasts brought to you by Karen Granddaddy and in this episode. We are covering the topic of communication within angular applications.Or communication within an angular type of ecosystem like react Redux.I’m going to talk about the framework level. Things also because most of the production applications. The web applications are generally built on Vue JS or angular or react so I’m going to take the popular choice. Ann just talk about how do we communicate between components between servers between clients and just? How do you go but passing data?Well, in Anglo if you see the fire base article that I have given you in the show notes and links one can do parent to child communication by the input tag and where what you do is you just define some square braces in the.Tag or the element the custom element that you want to send data into an the other. One will have annotation on Top of a variable call at the rate input and they receive data.The other way is you may ask how will the child communicate back to the parent for that there is something that is a concept called, is event emitter.And for that you need to attach an output property and all you have to do is pass it into the output property dot emit an you pass the object into that emit function and the other parent will receive it. Outside using the non square braces, which is like the curved braces so to pass data into a child you use input to pass data from the child?The parent you use output OK this is for simple communication and?The more easier way is by just injecting a service into both of them for example, if you just want to maintain the username or the user profile. You would just maintain a user service and simply use it as dot get dot store use angular properties. An you can use you can just use a global variable on a service an that’s the easiest thing to do.No, but I will also define something called, is a view child an in view child? What happens is you have reference to the.Do 2 to the child and you suppose if you are injecting something called as a child. A new element that you define within your HTML template. Then you can reference it by using something called as a view child annotation and then you can call anything of the child like the child dot something or child dot message child dot set something and you have access to the entire child elements.And uh.Yet another way of doing things like if you want a child to call the parents. Things then what you can do is in in the same input property that I had talked about in the first you pass on.You pass on you know a reference of something else like? How do you how do you reference an element in angular you just put the hash and suppose you have app sidebar you put hash sidebar and then the other one is supposed supposed navigation supposed navigation. Top horizontal line and you can just pass the sidebars reference into that and then one component can call the reference of some other component.The moose term logical thing to do is you know rely on a framework not use these bare bones things because very soon. You’ll get lossed, an which component is sharing information to another component an?It gets it gets very confusing so I would recommend you to use a frameworks like Angie RX where you define.You know a single story.A single store and.You know there is only one way communication only the reducers have capability to produce new store. You can’t just you know mutate the store. Just like that. You need to generate a new store and you don’t have to be terrified about you know, maintaining a single store. Becausr each each store can be subdivided within that OK. The store is supposed to be the single source of Truth, but logically you can.Separate them like you know that index dot JS file can refer to many elements so you can separate them by saying like OK. This is handling the stock ticker. This is handling the graph that shows the stock over 5 years. This is handling the news related to the company so you can maintain 56 toes and then reference all of them into the single store. Ann you use a framework to go about doing NGRX on Top of it?So NGRX is maintaining you know it should be a read only state an only if PR functions are supposed to update the state so this is in general. What redux is redux is nothing but you have a store, which has state an reducer is the only one that returns a new state and updates. The state and the action is the one that takes you to the radius and the reducer returns a new state so the UI.Can trigger actions like OK a button has been clicked?Nope, someone has opened a sidebar so you go and you go and start the UI keeps creating actions an you know reduces returns new state, but this paradigm is sometimes not enough. You you tend to have more complicated scenarios for example, on click click on often action. You want a server call to be made that just can’t be put into the reducer for that. You need to use something called’s effect.An actually before they affect came that unk so let’s talk about tongue and then I’ll talk about how come?Effects can simplify tongue so tank is nothing but a function that returns a function. So suppose if I say get so suppose if I say that get score of a particular game. It won’t instantly. Get me the score. It will return the function that calls the HTTP call over the get in the sense.Like that unk is a function that returns a function and this you can install a middleware and then your reducer and any action can become smarter by not just updating you know by creating a new immutable state why with the help of immutability GS or with the help of things like the spread operator. You’re not just limited to that you can also use thunk as a middleware to.Make these calls, but the easier thing to do is use something called his effects and effects is something that you use a method called, is create effect, an you pipe into the action an you tell OK when I get the ad book action, I will.Go ahead and make a service call and then once it’s successful. I will fire another action then that action can update the reducer so an effect can be fired that listens to actions and then it files another action on success or failure.This is one way or this is a way where long running operations in the.US redux is possible won the simplest way is thunk, but no sorry the complicated waste hunk and the simpler ways effects. So NGRX has a guide effect has a guy to effects. I’m putting it in the show notes.I end up.Most of the applications are riquier playing around with not just you know their data within themselves within you know communication. They also need to refer to server so for that. Of course, you can use a service rap it up with HTTP you can rap. The HTTP within a service. An yeah that service could be called from a chunk or from the effect.How however there are some more or scenarios for example, if suppose you have a fire and forget call or maybe a chat window. An you sometimes after 15 minutes or 20 minutes of server wants to talk.You you just you just can’t keep actively calling the server and asking are you done are you done. You need to use something more robust like websockets or even better use something called a signal are an.You can control.So web sockets is nothing but.Instead of you calling the server the server. Collins calls you and tells you that there is new data and you just have something in the UI that says on this message on suppose score. Update you do something around it and that also has an NGRX.Hum.Implementation where you just mention that OK in one of the effects. You mentioned that I’m interested in these kind of events like score update or job done an you can then fire new actions. From there and then that can correspondingly fire new effects or updater reducer. So NGRX has a has a good example over here called angular Contacts example, you can also use something.Like push notifications you can create a topic on SNS.Amazon services some is called as SNS, an you can create the same kind of push notification effects that the mobile apps have an Chrome also has similar mechanisms like fire base over.The Google push notifications or the APN S, an Chrome can also fire push notifications.It’s with the help of Amazon. SNS then signal are you would have heard about me talking about signal or signal is nothing but a better implementation of Websockets, where where web sockets is limited to just one kind of protocol. But sometimes you browsers are different, and you may not be having.The privilege to you know have web sockets at all types of browser. Then you know signal. I can fall back to some legacy means like service center events or if that’s not available. It will go back to long polling, but signal are abstracts that away from you and there is another library called Angie Oryx signaler, which can help you with the madness of handling updates so in that they have predefined some things like you know signal are connected.See disconnected state and they also have something like on message and there, you tell that OK, you maintain a switch statement where you convey what kind of messages. You are subscribed to and then from there. You can correspondingly fire actions and that can help you to get back the information into the redux system.There is another way of communications called, is indexed DB indexed. DB is the Chrome zone index database. It’s like a sequel built into Chrome an you could write make get calls get all Kohl’s. Find Kohl’s an makes equal kind of queries on the data that you have a simulated inside, indexed DB.I know or there’s a link. I’m giving you in the show notes that talks about how you do things like migrations. For example, OK, you’re opening the database for the first time? How do you populate data into it once you add a new column? How do you fill for the data that is already there, like make the previous columns is null and go about doing migrations or you could get by key get get all get by index you can add you can count.You can update you can aggregate you can delete you can do things like that on indexed DB and a significant way? How your user uses the application is.You need to have a mix of both online and offline scenarios, and for that. I recommend that you use something called a service worker so it’s nothing service worker is not a very complicated. Concept it’s a script that runs in the background. An they use the application cash to give the application and offline experience because your user is using your application in the.He’s on the subway train and then it goes through a tunnel where the network is not there, then it comes back then. It goes away and now people are in traffic there passing by different.Organizations that have signal Jammers sometimes uhm, so you need to have a good offline experience and service worker is a good way to abstract that away so that you know you can.Kevin offline experience an ensure that you can update the application.In a graceful manner or even continued amending the application while the user is disconnected.GRX also has a time to local storage. I have given one link local storage is nothing but something if you have files under fire me or sorry data and if I am bu can keep store storing things in a key value pair in so in the Chrome has something called his local storage and.Come there is Inter module communication you can achieve by using a shared module. I’ve given a link on how to go about doing that.For example, if one organization is made or something called as you know the billing module. The other organization will make the transactions module and you want a list of accounts to be synced between the two transactions and Bill payments because you know the from account will have the same list. Then you could use a shared module to do that, or of course, yeah, the NGO Arctic store is amazing to have that kind of information.No.What about those applications that are data heavy where you know?It’s You You have a canvas and you’re trying to imagine you’re trying to make a rule designer wear your shipping and logistics is happening that OK. I draw a human and then it. I draw like a container and how the human you have some interaction diagrams that show that OK human is going to load things onto the container and then the container will be put into a TU thatyou will be shipped.And then sent so you know this logistics train to have these complicated. UI is an hundreds of analysts are working on that UI.And you wish to coordinate stuff amongst them, an for that and sometimes you know, people send you know bar code scan bar codes or you know photos on WhatsApp that OK. This particular thing is accomplished an they take a photo of how it arrived so that you know the next party has faith that OK. The product is in a good condition, then you have to use the files.Um HTML5 files API, the file system is extremely rich you can keep.Maintaining things likeOh, you know you can maintain things like logs. You could maintain things like images. You can maintain complicated diagrams and sing between them. Becausr now file entry file system has given you an advantage to work with files directories in you know sandbox environment, so Chrome allows you access to a sandbox file system.And this enables you to there is it’s called as a file system and sometimes you need to request for more quota. If you’re using the web kit. If you want to handle things more than 1GB an the user has to click on grant permissions and now you have functions like onloaded. Remove ad with both files and directories and you can also deal with sub directories an have a so suppose I and you’re going to go about doing editing.Ann you have like a background uploader and you want things to be offline. An even a Mail application can become very complicated. With you know the users attachments. Sometimes you know the user clicks on have attachment offline. An just goes back to some other screen to handle something else. You need to be able to you know have like a download manager.To cash the data as in comes in so you need to be in a position to handle you know blobs an split.The files because some of your servers have 20 MB limits on file upload so you play with blobs and you don’t directly deal with everything so the file system. API of HTML is very mature take a look at it.I know if you are very bold enough you can also like put these of sometimes you need to make this uh. You know if suppose your your your web application being served is within a docker application.And being run from Cuban 80s, then.You may need to explore into things like world volumes. Persistent volumes an storage classes, so storage classes can help you abstract away things like as your file share over the blob storage OK. That’s getting 2 advanced but possibilities are endless. There are various ways of communicating with in angular across angular 2 angular from angular.Boot a two way communication one way communication simplex Multiplex duplex. Numerous scenarios are there. We have covered not just component component communication, but also things like reference passing thunk selectors?But then in getting into the code of redox you have things like.Effects I know the central point of store you have the socket. IO to have asynchronous updates signaler that is a higher level wrapper over sockets. IO there is a database type of environment using indexed DB you have the local storage where you can store the key value pairs, which are on fire. MB and then the entire file system is open to you make sure that you also leverage service workers to have a good offline experience.I know I’ll take a look at the files API. It’s capable of doing very complicated. Things your web you will be surprised to know a web application can actually replace your their desktop application.And all the best keep communicating.Keep um ensure that nothing is stagnant.The communication will ensure that the water body flows an it should not be stagnant like a Lake. Otherwise, it will start rotting let the reverse flow through channels of.Productivity andmitigate the fields of.Progress this is me current binary, signing off. I’m a developer and a technical lead at associated generally working in the innovation team. My LinkedIn is LinkedIn my user handle is in LinkedIn is KURTZACE my Twitter handle is Khurd says. And you could drop me an email at curves@aol.com so keep communicating keep the wheels turning.This is me current my daddy signing off bye.

 

– For creating new IBM account
bit.ly/QMeetup

– IBM Meetup
https://www.meetup.com/IBMDevConnect-Bangalore/events/266411519
Speaker: Shesha S Raghunathan – IBM Q Ambassdor

– Overview Quantum physics
https://www.youtube.com/watch?v=Usu9xZfabPM

– Experiment arena
https://quantum-computing.ibm.com/

– A Beginner’s Guide to Quantum Computing
https://www.youtube.com/watch?v=S52rxZG-zi0 and https://youtu.be/OWJCfOvochA

– Big picture Quantum Computing
https://app.pluralsight.com/library/courses/quantum-computing-big-picture

 

– Approximate Transcript of the session:

This is the Technology icing podcast and this episode is brought to you by Karen Bhandari, a full stack developer working at Societe Generale.
The content of this presentation is quantum computing. Most of the content material has been derived from the conference – I had recently attended on the 7th December 2019 at IBM campus on IBM Q as well as a few presentations and materials- I have come across at brilliant.org as well as plural sight quantum computing, the big picture and QES kit-Curated by IBM now.I may not be an expert commenting on.The developments of quantum computing. It is just some random.net angular sequel. Some beginner and machine learning guy is trying to figure out what is quantum computing like when he attended the conference or me? how I feel about it and what kind of differences? do I see from a full stack perspective?So don’t consider me to be a physics expert expert becaus my only knowledge has been till 12. Standard and I’ll do my best on conveying what I’ve understood so far and hopefully someone in the community could correct me.Quantum computing.Let’s describe this as something which is opposed to what opposed to classical computing. What is classical computer where I’m recording this podcast on a classical computer something?That is a laptop an I have a smartphone with me which I carry that’s a classical computer.And these are generally manipulated by bits, which is generally.In either or state of 0 or one like a light switch.There is another realm of computing, which people are trying to explore called the neurotic computing like they’re trying to emulate what the brain could do.And they are trying to use signal theory to see how one cell of a nerve is axeon dendrites are trying to send signals to the other nerves. Anne, how a memory based learning happens like bit by bit is bit by bit logic is something like.If I need to represent 2 it will be.Of one zero if I represent 3. It’ll be 1 one if I represent 7. It will be 1. One one, so I’m representing the knowledge in bits.In a neuron.Oh do you have?Um.They have something called as weights and biases.In the sense, I’m learning match by doing memory for example, if I want to do mathematics on multiplication. I don’t do bitwise computation. I just remember the entire.Table mathematics table for example, till 20 and then I or maybe even 10 is enough and then I extend this to tables with the concept to other two other greater tables with the with the concept of carry over, so I’m I’m trying to do mathematics by memory. This is neuron type computing an the 3rd part which I’m going to talk about is the qubits now cubits.Can be imagined to be something like?Um.The regulator of your fan or your thermostat where you could you could be on or off but you can also be somewhere in between for example, OK, you could be either at.Oh, 45 to 50 Celsius or you could be at 5 degrees Celsius. An you can also be somewhere. In between with the help of regulator so qubit could exist in more states than a bit could exist in an this actually comes from quantum physics, and quantum physics deals with mostly amplitude and the probability distribution of Electrons.An quite a lot of famous physicians have tried to study the phenomenon of light being both or particle as well as a wave form so.If you treat a wave like if you’ve seen a wave like in music or when some of your friends have been connected to heart rate monitor or you could see waves to be something that goes up and down an there is something called’s Crest and there is a tough. The bottom part which dips down and the upper part is crashed. So we say that an electron exists in a particular area. If there is a crash that is there is some mountain.In the waveform.And in the absence of it is denoted by the bottom part of the wave.In in the case where electron if we need to find out what is the state of the electron we find out the position by finding out the position of the upper part and?Uh some people have also tried doing experiments if you research and Google something called as a 2 slit experiment.It is nothing but you take a paper and you make 2 slits in it through with the help of a scissor vertically of both slits are spaced by a few millimeters and then when you flash your paint your paint gun around it. You’ll see that OK your paint is you know, and if you put another surface behind these 2 slits you’ll see that OK. The paint gun occupies the parts where there is gaps.But in case of electron gun.Uh the if suppose you have an electron gun and you shoot waves.And if they go through the slit and so some ways goes through one slit another waves goes to others. Let’s then they form. These interference patterns and they don’t form 2 slits they form multiple you know multiple marks of paint of electrons everywhere. Spaced by a few did unit distances apart an this interference is something that quantum computers.Try to leverage or to find out what is the state of the electron an they use probability distribution for example.Probability distribution is something that is used in logistics regression where you trying to classify whether it’s a cat or a dog or whether it’s or human or an animal or this? What they do is they make something called AZS curve. I think that’s getting into too many details.IBM presented to us.The IBM Q System One and this is the quantum system. One Anne they tried their best to maintain this cubits at minus 273, – 270 degrees Celsius, which is one form of supercomputing called the superconducting super superconducting quantum computing.And superconducting quantum computing is needed becaus I think you want to stabilize the Electrons because I have a time. Not very sure, but this is what I’m feeling that if the electrons are random in nature. Then it’s very difficult to find out the position becaus they’re having momentum.And Uh Heisenberg has a principle that states that if you want to find out the position of the electron then you should not have any momentum and if you have position. You cannot find out the momentum that is.You.You you should not be able to find out the wavelength. If you knew the amplitude of the wave and you can’t find both at the same time you can’t have a wife and you can’t have a girlfriend at the same time, OK sorry bad analogy.And so they say that so that so I think that’s why they try to cool, it down an Deep Throat right OK in a normal conduct of what happens is the electrons flow from one end of the conductor to the other end or in the magnet decree 2 pools. North Pole and South Pole and the electrons are oriented in a specific fashion so electricity can flow through.But in this case, they don’t make things flow at the artist stationary. The Electrons are stationary and based on the orientation defined out? What is the qubit state is?So IBM Q System one. UM, I I saw the the photos. They were trying to show they had some dilution refrigerator and the qubits are actually they have just 50 cubits in the largest computer but 50 cubits is enough to simulate or is enough to have like the all the items or not all the items like a good chunk of more than 50% of the items that that are present in mother Earth, 50 cubits can represent that much information.Because they say that OK, the bits if you need to represent.Eat you need to have 3 bits like one or no, yeah, so you need to have 4 bit sorry and if you need to represent 7. You need to have 3 bits and then the bits keeps growing at 2 to the power of something but in the case of quantum.The theso your your two year 2 bit skin represent more than so around 4:00 and then you’re 3 can represent 8 and much more so. So, your cubits if you have something like 50 cubits. I think you can have like trillions and trillions of combinations because they’re growing at such exponential levels.Us so that’s why they have it at the size of farm which is good enough for you know decent simulations exercises.They have a dilution or refrigerator, an what happens is as and when you go from the Top to the bottom they try to reduce the temperature further further further photo for example, above is 40. Kelvin Kelvin is nothing but like if it’s 1 Kelvin. It’s know if it’s 0 Kelvin. It’s 273 degrees Fahrenheit in Celsius and.So should what do you trying to keep the chip is in something like 10 to the 10 to 15 Millie kelvins and what did what this does is they have an in. They have a classical computer and a supercomputer so this quantum computer mixed together, because you need.The wave of the wave system to communicate things to the classical computing becaus. It typically you would not have stat kind of superconducting thing in your own house. But what IBM claims to have done is they have made it in the size of a data center and they have this massive cooling capacities and it typically people would submit web jobs from a desktop computer so you need to translate your.We’ve type output to the normal output so they interface microwave signals an they can resend it to an FPGA and this and they have a big kind of enclosure to prevent noise from coming in, so in this noise is not nothing but it’s not the sound noise, but quite a lot of things can be noise for them. For example, even if the temperature increases the hyperactivity of the electrons increase, and they start releasing.Random photons an so that’s why they need to keep it at super cool temperatures. This one move. There are many movies that quantum computing is set to have developed but I think most of them are theoretical. Most practical things I may have seen or come across I don’t know is is you know in the superconducting side like for example, there is one more type of of.Computer one more type of quantum computer there. Actually, 3 other types, which you will get to hear in the plural sight course and you see you see Microsoft. Google IBM have started investing.Or heavily into this realm, and this is very promising becaus.Dinner just have potential to solve things that classical computer could not have, but can it can also unlock other path breaking things other part breaking scenarios that we may not have thought about.So.Uh let me give you some examples of where this quantum computer could be used. I’m sorry if I’ve mentioned super computing in the?Pre previous few minutes because I feel that quantum computing can be close to supercomputing becaus of its capabilities.US first one of these chemical simulation for example.When you have molecules with you the scaling of the molecules become very exponential.Oh, because of a single molecule is consisting of many atoms and atoms are having.Protons neutrons and electrons orbiting around that nucleus, which has that and a simple thing like caffeine.You know caffeine or simulation can take a numerous like it takes more than 20% of the items of the earth or or 20% of the atoms of the earth that many times of bits to simulated which is actually Humongous. So so I feel that if you want to simulate caffeine molecule. It’s almost impossible. I’m saying that caffeine is is not that simple, it’s a hydrocarbon.An you may feel that OK. It has just 16 nodes in it, and why you saying it takes so much.Even if you even if you’re trying to arrange your friends your 10 friends on a single dining table. You have 10 factorial combinations, which is a very big number so even a simple thing like.Oh caffeine simulation or maybe if you want to discover new drugs how how how can a drug latch onto a virus and inhibit you know it’s effect to cause cancer to you an you know to simulate so many things build a compound. It’s it’s it takes a very long time you need to also perform simulations and accuracy in simulations is generally 50%, so drug discovery is something that is very important in the other one is battery research.Uh because the you need to find out the smallest molecule that can store more energy reduces size. You need to simulate things at molecular level. Then the protein structure material Sciences gene analysis. There are so many things that have exponential types of computing and in artificial intelligence have seen that in CNN, which is convoluted neural networks.This sample the images to Dumbo things to find out what it is for example, if they’re looking at.Uh something like a water bottle or let’s let’s take for example, or human with eyeglasses. They just convert the eyeglasses 22 circles and they you know sample just outlines and they dumb down in the sense your alphabet a has just got one horizontal rod and then one or we which is inverted upside down on the very small Top and just 2 lines, which are 45 degrees.And then you know they try to sample down the images by a big margin so an most of them operate on 64 by 64 pixels to find out what the image is there be cause. If you have to find out the combination of all the pixels, though it will be very difficult for a normal classical computer to do so even in CNN. The part breaking image classification algorithms try to sample it down so that it can be tractable by a normal computer. Even the NP hard problems or the simulation experiments like Monte Carlo.Predictions or risk analysis of portfolios these could be solved by quantum computing and then.Uh IBM’s claims that if you can convert anything to a matrix they can take it and you know perform some things to enable those quantum computer to understand these instructions and.In this eml also they did not do so much approximation and they can do more convergence on the dimensions and better classification. You you.You can even go to an advanced level like for example, you need not allowed will to be classified as a fish because it is a mammal it is not it may be a fish it cannot be official is still a mammal which can swim.Oh, but but if you do, but you know if you need to make a CNN understand that OK. It’s avail. You need to give it move it more data like OK, not just swimming not just being in water, but also OK. How does it give birth an? How does it reproduce so these complicated you know you in AI? You start giving more access of data is an I think it becomes more and more difficult as in when you add too many things to be considered becaus nature is actually quite random.It’s not very predictable anything can happen to anyone like oil.Something like a butterfly flapping its wings in Brazil may cause a tornado somewhere else. I don’t know like the people. That’s called the butterfly effect, so nature is very unpredictable an?DSC not up the the quantum bit is something that gives approximate results. It does not give the exact results. Becaus sometimes noise comes into place. So 2 + 3 becomes 4.9999 and then you apply some probabilistic layered over it, saying that OK 2 + 3 becomes 5. So if IBM advice that do not use quantum computing for things that classical computer can solve.Um Ann.Oh yeah, I just received the I just remembered what you need 10 to the power of 48 bits in classical computer to handle or simulator caffeine molecule OK. Let’s get back to the the computation part so.The classical thing is measured by how many flops or teraflops are there, but in quantum. They are measuring by the error rate because you know noise like temperature interference bad coherence an calibration crosstalk. You know the wave issues that you see in the telecom industry can also start affecting this an and the the characterize the error rates.And and you know becauses 2 + 3 gives you a probabilistic answer of 5. They say the more number of times you execute the answer. You need to see like I draw? What probability the answers coming so you try to get many you try to replicate it many times and see. Where is the probabilistic answer going towards so you won’t get a direct answer so and of course as I told you before don’t try to do things that classical computer can do like addition is very simple.But complicated cases like simulation or you know trying to put in like 10,000 variables. These are something that where the state space is very high where the state space becomes exponential, then you should try using the quantum computing an?The noises should be reduced in the sense make sure that the quantum. Computers is at a very optimal temperature that the cube. It should be super cold and for for example, I know I think there is other types of so other types of quantum computing something that Microsoft talks about frequently is called a topological ones.And then there is spin. State then there is light polarizations or different kinds of companies are experimenting IBM claims at Microsoft is still a theoretical stage.And know what Microsoft has done for the community of quantum science is they have given them the quantum development kit and you can write Q Sharp and you can simulate quantum commands to be given to if the quantum computer vertebrae invented by them, but IBM has given you know the entire community. You know, some computers at their disposal and you can also you.Normal man can also submit jobs to the quantum computer today, so most of their knowledge is located in qskit.org an.That’s the training and the community involvement, an where people go to submit their quantum computing jobs is quantum-computing.ibm.com. You need to register with an IBM account and they have 2 kinds of ways, you can write quantum codes. One is a circuit composer.And the other one is QIS kit notebook, which is a quiz kid. Notebook is very similar to a Python style notebook and circuit computer is where you drag and drop gate. San measuring things, then you connect them together, something like a digit like an online breadboard. They give you and you drag and drop the items, so those we see the big inning is use the circuit composer.Uh to do basic experiments an cues kit has some impressive examples for if you go to the quantum computing. Jupiter notebook tutorial. They have a PY&B notebook jupyter notebook that has shown you how to do stochastic vector modeling an they have shown you? How do you go about doing?Do I add the adversarial network GNN uhm so generating adversarial network? Is nothing but a good dog bad dog like if the dog does the right thing you see Oh well done and if the dog does not do the right thing you say bad dog and you don’t give them the biscuit?And do you have a so so they are exposing their machine learning libraries through?A library called, is Akwa.An they have other things that they are offering like you ice skate chemistry. QoS kit finance guy skit AI that helps you to go about doing.We’re trying it out an what happens over here is once you submit a job. You need to select one of those freely available quantum computing machines that they have given to the community. So when you submit a job it gets queued in the system. For example, if you are 16th in the queue. Then, after 16 jobs are finished then you’re you get a chance to execute your code.So they have won super 1 quantum computer in Melbourne in Essex in Burlington in London, an QSM simulator and each most of them on an average have around 5 to 20 cubits. An these are the backends available to the community and they can submit quantum jobs, either through the notebook or the circuit composer.Or through quantum-computing.ibm.com and can see the results of the same.It’s a fairly complicated topic. Microsoft has also invested in Q SDK which you can run simulations on your classical computer to an so when the quantum computing is made commercial then you may be able to start using the same instruction set as Microsoft.QSDK and You will be able to get exposed to how to do programming against it. An yeah, I think IBM. One is available for the community and recently in a WS reinvent the OR have also shown how to go about doing quantum computing an eye. I strongly recommend you to look at the YouTube video and some of the foundations of what is quantum physics and the quantum computing big picture?Included site.Uh also explains like Barebone idea of what quantum computing is about an IBM has excellent resources in the quest kit just ecosystem that they have so start experimenting and stay sharp stay relevant. This is Karan Bhandari signing off have a productive week ahead or if you’re listening in the weekend have a pleasant weekend goodbye.