Sargon Benjamin

Cracking the CCAT: Spotting Patterns in Logic and Number Series (Part 4 of 5)

2026-05-02T21:20:00.000-07:00

The logic section of the Criteria Cognitive Aptitude Test rewards one specific skill above all others: pattern recognition under time pressure. You have roughly 18 seconds per question across the 50-question, 15-minute test, and logic items don’t give you anything to fall back on — no formula sheet, no vocabulary trick, just sequences of numbers, letters, or shapes that follow a rule you have to discover fast.

Here are the patterns that come up most often, and what to check first.

Arithmetic series. The simplest case: each term differs by a constant. 4, 9, 14, 19, ? — add 5. If the differences aren’t constant, check whether they form a series. 2, 5, 11, 23, ? differs by 3, 6, 12 — the differences are doubling. You’re now solving a meta-pattern.

Multiplicative and exponential series. 3, 6, 12, 24, ? doubles. 2, 4, 16, 256, ? squares the previous term. When numbers grow fast, multiplication or exponentiation is in play — never try to force addition on a runaway sequence.

Alternating series. 8, 3, 10, 6, 12, 9, ? — two interleaved sequences (8, 10, 12 and 3, 6, 9). Whenever a series looks chaotic, split it into odd-indexed and even-indexed terms before giving up.

Letter series. Convert letters to their position (A=1, B=2…) and the same arithmetic rules apply. B, E, I, N, ? skips +3, +4, +5 → next is +6 → T.

The 5-second rule. If you can’t see the pattern in five seconds, try the differences. If differences don’t help, check the ratios. If ratios don’t help, split into alternating subsequences. If none of those work, mark a guess and move on. Every logic question is worth the same as every math question, and burning 90 seconds on one is how strong test-takers run out of clock.

Speed in this section comes from drilling, not from being clever in the moment. After 30 timed items, the patterns start to feel like familiar shapes instead of puzzles.

Practice timed CCAT logic questions on TestCCAT →

Up next — Part 5: The Test-Day Game Plan.

Is Stamina Decay holding back your CCAT score?

2026-04-22T11:55:00.000-07:00

If you recently took a practice run on TestCCAT, I wanted to follow up on the metrics you saw in your report.

Most candidates who miss the cut for high-stakes roles at firms like Coinbase, Crossover, or Vista Equity Partners don't fail because they lack the logic. They fail because of cognitive latency. The CCAT isn't just a math test; it's a 15-minute high-pressure sprint.

If you noticed a dip in your accuracy during the final few questions of your practice test, you're likely experiencing Stamina Decay.

My advice: Use our free simulator to practice the '5-Second Pass.' If you can’t identify the logic pattern in 5 seconds, skip it. You can practice this as many times as you like for free on our homepage, or check out our full logic breakdown in the Exam Prep Guide.

Upgrading to TestCCAT Pro gives you the engineering-grade tools to fix stamina decay:

450+ high fidelity replicas: Practice with questions that match the 2026 proctored exam cycles.
- "The math question felt very similar to the ones in TestCCAT"
- "your CCAT test product is great. Very helpful and the questions were pretty similar in difficulty to Criteria’s questions"
Tutor insights dashboard and projected score: See where you're losing focus and which specific questions are time-wasters for you.
Category specific drilling: Drill only the Spatial, Math, Logic, or Verbal patterns where you're weakest.

Don't leave your next interview stage to chance.

Cracking the Criteria Cognitive Aptitude Test: What to choose when you don't know the word (Part 3 of 5)

2026-04-17T09:25:00.000-07:00

When you hit a synonym or antonym question and the lead word looks like total gibberish, the most effective hack is to stop looking at the question and focus solely on the choices. Most CCAT verbal items are structured so that four out of the five options share a similar theme or linguistic root. By identifying this "cluster" of synonyms, you can instantly eliminate the majority of the answers without ever needing to define the target word.

The correct answer is almost always the outlier that refuses to fit into the established group. This logic-first approach turns a vocabulary test into a pattern recognition game, significantly boosting your accuracy even when your vocabulary fails you. You can find the full step-by-step breakdown of this clustering strategy and other verbal shortcuts in the "Verbal Reasoning" section on testccat.com .

How are recruiters spotting real talent in today's AI arms race?

2026-04-17T08:24:00.000-07:00

Is it just me, or has hiring turned into an AI vs. AI arms race where the human element is getting lost in the shuffle?

I’m genuinely curious how recruiters are spotting real talent now that keyword-stuffed resumes and AI-assisted tests have become the new baseline. On the flip side, are candidates finding that these tools actually help them stand out, or are they just a survival tactic to get past the bots?

How are you navigating this "broken" system to find—or be—a genuine hire these days? I’d love to hear your thoughts on whether we can actually fix this, or if the old way of hiring is gone for good.

I’m curious if resources like TestCCAT—a proctored study guide that allows candidates to practice for their employment screeners so as not to get flagged for cheating—can help restore some integrity to the process.

Cracking the CCAT: Spatial Reasoning with real examples (Part 2 of 5)

2026-03-31T08:12:00.000-07:00

On the Criteria Cognitive Aptitude Test, the spatial reasoning questions are a high-speed measure of how quickly you can mentally rotate / manipulate objects and detect patterns. What many candidates don't realize is that these visual patterns often form the core of the "Proctor Integrity Check"—the high-stakes follow-up assessment companies use to verify your score in a controlled environment. To help demystify this, I’ve put together a guide featuring actual CCAT Spatial Reasoning examples that I encountered during my own prep. Mastering these specific 3D rotations and matrix patterns is the best way to ensure your initial score holds up under the scrutiny of a proctored re-test, turning a potential "integrity flag" into a confirmed job / program offer.

Cracking the Criteria Cognitive Aptitude Test: The Math Cheat Sheet (Part 1 of 5)

2026-03-27T07:48:00.000-07:00

A few months ago, shortly after my five-year tenure at AWS, I found myself staring at a "Thanks, but no thanks" email for a role I knew I could do in my sleep. The culprit? A failed CCAT. As a software development engineer & manager, it stung to be filtered out by a 15-minute timer before I even got to talk shop with a recruiter. I realized then that the CCAT isn't a measure of your engineering talent—it’s a high-speed "rhythm" game where 18 seconds is all you get per question. I spent the following weeks reverse-engineering that pace to build testccat.com, and to kick off this new series. I’m sharing my CCAT Math Shortcuts Cheat Sheet. These are the exact mental hacks I developed to stop wasting time on manual calculations and start clearing the 50-question gauntlet with confidence. Reclaim your time and focus on hitting those elite 90th-percentile scores required by companies like Vista Equity and Coinbase

It says it is free and the book is not a good one

2019-05-19T00:28:00.002-07:00

"It says it is free and the book is not a good one". One of the most interesting customer responses I've received
https://www.pushbeta.com/2019/05/19/it-says-it-is-free-and-the-book-is-not-a-good-one/

Peloton founder John Foley on podcast ‘How I Built This’

2019-05-06T10:08:00.002-07:00

Really enjoyed listening to Guy Raz interview Peloton founder John Foley on the “How I Built This”podcast. Really interesting how he likened the journey of the video game arcade transitioning from a public venue and moving into people’s living rooms with the the boutique fitness class movement.

I listened to the episode while working out and found myself grinning and laughing multiple times at some of John’s remarks. Definitely going to listen to this one again – numerous wisdom gems to learn from!

Stripping the layers – how secure is your docker image?

2019-04-09T23:28:00.002-07:00

Check out my latest post on potential vulnerabilities in intermediate docker image layers

Seeding mongodb via docker-compose

2019-04-07T17:15:00.003-07:00

I recently wrote an article on my blog about Seeding mongodb via docker-compose

Data Flow for sargonsays.com English to Assyrian Dictionary

2019-02-01T11:17:00.003-08:00

The most challenging and interesting aspect in designing an English to Assyrian dictionary is in ensuring data integrity. What does that even mean? Here’s a straightforward example… When you search for something like the color “red”, you expect definitions to appear in a ranked order matching your intent. You expect to see results (Assyrian words) whose definitions are the color “red” and then a ranked list in descending order of related terms like “reddish” or “red rose”. You don’t expect to see words like “shREDded” which contain the letters “red” but don’t actually match your intent.
Here’s a glimpse into how this is made possible.

End Result

This is made possible thanks to String similar algorithms, lucene search indexing for scoring, MongoDB text search index, and basic String heuristics to ‘guess’ at the most important words in a definition sentence.

Related Terms

Another useful feature is the ability to suggest related terms. Maybe the user mistyped what they were looking for. Mongodb has a really nice text search feature that makes this possible

Great podcast episode using mental models to make better decisions

2019-01-28T14:21:00.002-08:00

The Art of Manliness podcast recently aired a thought process / decision making episode, interviewing former Canadian intelligence officer and owner of Farnam Street, Shane Parrish.

The 5 key takeways are crystal clear on the principles page of the Farnam Street blog. Do check it out: https://fs.blog/principles/

I've shared it with my team at MapR and highly encourage others to take a look

Learn Kubernetes by doing

2019-01-20T16:16:00.002-08:00

Lots of learning styles out there. I prefer to learn by practice. Getting my hands dirty and going from A -> Z, quite often using the straw-man approach.

My employer, MapR, is working on a native Kubernetes (K8S) offering. To this end, understanding how to use K8S via a simple app seemed like the right place to start onboarding and socializing other engineers to the K8S ecosystem.

Arun Gupta of Amazon created a great tutorial that goes through the following:

Compile run a java web app locally via maven
Then run the same app using Docker via a multi stage Docker file
1. He includes customizations for shrinking the Docker image (I skipped this part)
Deploy the app using K8S
Deploy the app using Helm to simply the deployment on Kubernetes
Deploy cluster on cloud K8S provider and change kubectl contexts to point to different clusters
Istio and Traffic Shifting
1. I skipped the Istio portion and just followed Traffic Shifting. Very useful exercise here

I forked Arun’s repo to suit our needs. That repo can be found here

Having fun remotely

2019-01-20T10:38:00.002-08:00

How does your entire team have fun together? For innovation to breed, there needs to be a safety net that nurtures the sharing of ideas. The sharing of ideas must be a safe experience. Team members need to feel that this isn't a popularity contest and that even the most insane and silly sounding ideas should be brought to the table, because those ideas can trigger a spark that lights up another idea.

One of the ways I promote a stronger safety culture is by giving time and listening to the creative process. A way to enable this is by incorporating silliness over seriousness. Design Thinking does a great job of this. An exercise I really like from Design Thinking is by asking the question, "How might we come up with a solution if this were a Disney movie and we were characters in it?" Yes, I want my team of software engineers to put themselves in a Disney animated film and imagine up a solution so wild it may be fitting for a Disney film. Some Jenkins testing pipeline that applies a patch not so different than building Mr. Potato Head in a Toy Story scene or a mapping solution that turns Aladdin's carpet in to a live GPS map to help him navigate the Arabian peninsula. Think silly and encourage it.

Now, I couldn't just show up to a team meeting and say, "Ok everyone, let's act silly". Well, I technically could and that would be part of the process but it's not an end all. It must be encouraged and reinvigorated daily. The creative process is a build up of a series of daily micro events that promotes the culture of stronger safety. Laughter helps. Even in crunchtime.
Sounds easy and is indeed, that's the good news. The challenge comes with doing this in a remote work environment.

In a recent team meeting, we remotely played a game called Jackbox Party Pack 3. It made us feel like kids. There were side jokes, jokes really only understood if you were part of a geo region, and all around good fun. It's no holds barred when even your Director of Engineering has to come up with a creative wrestling move that has the word "chunk" in it! Jackbox is a collection of mini games that put your emotional IQ, creative thinking, and quick wit to the test. In some ways, I liken it to a video game version of the NPR game shows "Wait wait, don't tell me!" and "Ask me another".

My team is decentralized. We're in 3 different geolocations with 3 different timezones. North America, Europe, and Asia. Getting everyone together for a team meeting itself requires that team members perform yoga stretching on our schedules. Webex and slack really do bridge the physical distance between team members and for those tools, I'm thankful. Yes, notification nightmares do happen but the tools serve their purpose and allow for customizing notification policies to the nth degree.

Remotely playing Jackbox Party Pack3 with 3 different geolocations was one of the first times my entire team has had live fun together. Silliness over seriousness as a means of encouraging creativity by providing a stronger safety culture. I'm not paid to promote the Jackbox product. In fact, I have a vested interest to not promote it so that I keep the upper hand at this game during family and friend get togethers, where we'll often play one of the Jackbox mini games!

Hackernews has a post about Assyrian king Ashurbanipal

2018-11-22T10:46:00.001-08:00

The hackernews post can be found here https://news.ycombinator.com/item?id=18508530

Hiring for my team at MapR - Senior Staff Software QA Engineer

2018-10-18T23:13:00.003-07:00

Interested in Big Data, test automation, and excited about learning on the job? Do the words Docker and Kubernetes tickle your fancy? If so, then maybe this is for you.

http://jobs.jobvite.com/mapr/job/oEkw8fwm

Silicon Valley Code Camp

2018-10-17T22:28:00.003-07:00

This last weekend, I attended the Silicon Valley Code Camp hosted at Paypal in San Jose. Aside from being 3 minutes away from the gym (this is a MAJOR factor me), the session schedule and speaker lineup looked solid. In particular, what caught my eye were the Technical Leadership / Management track and the cloud/container/kubernetes focus.

My tentative schedule looked like this:

Saturday
9:30am -mgmt: managing in an agile world
1:00pm - mgmt: leadership lessons People learn the hard way
2:15pm - mgmt: you don't know what you don't know! insights
OR
2:15pm - k8s: Java devoloper's journey in k8s land (easy to follow github repo instructions . Really good session)
3:45pm - 400 Billion messages a day with Kafka at Paypal
5pm : top 10 microservices with Redis

Sunday
9:15am : container: Terraform + Packer + Docker = winning devops
10:45 am Sunday: k8s: redis in containers and k8s
12:45pm - MongoDB aggregation like a champ : AWESOME session by Nuri Halperin (this guy is hilarious too! Also helps that he knows what he's talking about. Was quickly able to cook up some queries and code ad-hoc to address questions from the audience)
2:15pm - help them stay : leading for staff retention
OR
2:15pm - How to ruin any software development project

Definitely going again next year and highly recommend checking it out!

First article on LinkedIn

2018-05-08T19:58:00.001-07:00

I published an article on LinkedIn

What we learned about quality from scenes in movie Deepwater Horizon

I'll still write blogs here. Here's the article text as it is on LinkedIn.

Viewpoints regarding the factuality of the Deepwater Horizon adaptation aside, assessing the tragic 2010 oil spill, which claimed 11 lives and has lasting environment impacts, makes for an intriguing QA case study.

Relentlessly stick to your core values

I lead a team of Software Engineers in Test at MapR. To reunite with the core values and integrity that is required to ensure quality, we did something different in our most recent meeting. Rather than go over schedules, cool things we’re building, or tradeoffs with particular design patterns in our approach, we self chartered a quality case study of the tragic 2010 oil spill chronicled in the movie Deepwater Horizon. This involved watching scene snippets and steering an open-ended discussion after each scene.

Clear Communication is key

The film is ripe, even juicy, with concepts that will tickle a quality oriented engineer from a non-technical standpoint. These are people issues and pressures, which can be tougher to navigate for someone who really values hard-core technical skills.
Corporate villainy, the pressures of schedule and deadline misses, cost cutting, subversive fear tactics, and omissions of truth both wreak and linger in the film. Some subtle, some very subtle, others not so much.
Procedural concepts like integration testing, negative testing, and questionable test results add to the film’s legitimacy from a quality engineering standpoint.

“Hope is not a tactic”.

It was interesting to hear some team members really internalize the struggle and pressures that come from ensuring quality while not killing the buzz and camaraderie of the entire project.
Others were quick to discount the entire movie, because the process followed by the teams in Deepwater Horizon was deeply flawed and lacking any engineering discipline… Where were the checklists, where was the acceptance criteria, what was the fallback plan for having questionable test results, where was the accountability matrix!?

Listen all the time and leverage your experience

My take-away... In the absence of checklists and acceptance criteria, rely on your listening skills and on your experience. Listen to that which is unsaid or omitted. Your experience will help you detect failure before it occurs. This usually takes no shape; rather, it is unstructured and may come to you in the form of a bad feeling or gut reaction to something omitted. Listen to your twinge, then act on it.

If you’d like to run a case study with your team using the movie, I’m including the order of scenes we watched & discussed:

Set the context…
Deepwater Horizon, a semi-submersible oil drilling rig floating 5,200 feet over the gulf floor, 42 miles off the Louisiana coast, experiences blasts and explosions causing a massive oil spill. 128 crew members on board. 11 members died and millions of gallons of oil spilt across the Gulf of Mexico, BP settlement punishment of $21 billion. Environmental impact, impact to human life, and a real financial and PR impact to a corporation.

Scenes:
1:21:20 -> 1:22:00 Emergency
(havoc, emergency, devastation, fire, evacuating on rafts).
No discussion here necessary. Just setting the brevity of the situation, starting with the devastation and tragic impact first. This sets the tone for the team to start looking for cues early on during our review of the scenes and eventual discussion of course correction.

15:37 -> 17:20 Crew Arrives
(The Transocean rig operating crew arrives on a helicopter and finds out that BP tells the test team to go home. Hard for the team to hear what’s going on because of the helicopter noise. Commotion about Schlumberger team not running some cement test. Cement was put in by Haliburton)
Ask the team for open-ended thoughts and feedback.

28:10 -> 31:45 Confrontation
(The Transocean rig manager confronts and berates BP for not performing the cement bond test. In its stead, he mandates that a negative pressure test be performed as a substitute test.)
Ask the team for open-ended thoughts and feedback.

33:00 -> 35:07 Side Conversation
(BP exec played by John Malkovich bypasses the chain of command and talks big picture with Transocean crew member. Crew member tells him a story about fishing, sticking his hand down a hole but plans for getting bitten, not hoping. Hope is not a tactic)
Ask the team for open-ended thoughts and feedback.

35:30 -> 47:40 The Test
(Test is inconclusive. Followed by a whiteboard session regarding technical nuanced details of a bladder effect, to which one engineer says it’s a little above his pay grade. Conversations floating around regarding test inconsistency due to potential problem with drill pipe or sensor giving a false reading or any other of 1000 parts moving down there)
Ask the team thought for open-ended thoughts and feedback.

1:24:20 -> 1:28:00 Hero
Ask the team thought for open-ended thoughts and feedback regarding teamwork or lack thereof in this scene.
---
Relentlessly stick to your core values, Clear Communication is key, hope is not a tactic, listen all the time and leverage your experience

Data to the Rescue for SargonSays dictionary

2017-12-26T15:56:00.002-08:00

Anonymized trend data is tracked on the SargonSays site so that we can create a better user experience.

Happy users = more usage of the dictionary = keeping the Assyrian language alive.

Data Analytics: Users aren't using SargonSays solely as a dictionary; rather, they are using it as a translator.

People are typing in long phrases and expecting the site to respond with results similar to Google Translate, which doesn't yet have the Assyrian language. Rather than discourage users from using SargonSays as a translation tool by giving them no search results, why not make them happy? Why not attempt to help them in a meaningful way?

Starting today, the site will provide users with a set of "Related Searches" that best match their search. So even though we can't yet provide you with a word-for-word translation when given an input phrase or sentence, we'll atleast be able to help users stitch the words together, like stepping stones.

This was made possible via a text index search feature in MongoDB. The data in SargonSays is pre-processed so that searches are fast; however, this means that real-time fuzziness is non existent. SargonSays is built on the premise of being an extremely high quality dictionary with high fidelity search results for exact words, not long phrases.
Huh? To accommodate users who are looking to use SargonSays as an English-to-Assyrian translator, I made use of a feature in MongoDB that allows for free form text search by indexing certain fields and placing weights on the more important fields.
The end result is that users will now get meaningful search results when, previously they may have gotten nothing back at all.

I'm currently only indexing a small part of the dictionary - just the term relationships. This provides users with high fidelity search results and a simple user experience. Next steps would be to give users the option to fully control an advanced search experience. Here, I'll remove the training wheels and expose a full text search of the underlying content itself, rather than just the relationships. Dictionary definitions and meanings that normally wouldn't be exposed will now have a high chance of being unearthed.

Aveeton basimeh (may you be thanked!)

UI and API continuous testing monitors

2017-02-20T20:23:00.000-08:00

Do you cron your UI and API tests pointing to production using your own test clients or do you use a hosted solution?

I've been looking into the following:

New Relic Synthetics

overall, seems like a simple and quick way to get some basic API (using the vanilla js Request lib) and UI (webdriver js) monitor tests up and running

only supports Chrome web browser

Set up a CI trigger to run specific tests:

UI tests on SauceLabs or BrowserStack

support for mobile and browser configs

API tests from the CI agent node itself

We're currently using the latter option as it grants greater flexibility and robust support for various configurations.

Though, the added nicety, aside from the simplicity, of New Relic Synthetics is that it unifies the testing codebase to javascript for both the UI and API tests. Which bears the question, should your API and UI tests be written in the same language? Sure, they can both be BDD, but is it important that both the API and UI tests be written in the same language?

API Test Framework boilerplate in Java

2017-02-09T20:03:00.002-08:00

I created a repo on github for a starter boilerplate framework to build off of when doing automated API testing.

language: java
test framework: testng
http client: Retrofit & OKHTTP by Square
logging: logback / slf4j
build & dependency management: maven

I'll passively update the repo as the project is meant to be barebones and for anyone to pick it up and fork / modify it to suit their needs.

Some cool features it supports:

an http interceptor (okhttp by square) that allows you to do things like inject the headers and various url params before sending a request as well as logging both the request & response without each test having to do so
each test method gets its own log file thanks to the logback MDC feature
each test has fine grain access to the http request and response

You can find the repo here

Most frameworks have architecture diagrams so I'm including one as well...
🔵🔶🔷🔴⬛⬜😇💪😆
...
I hope that diagram was helpful

English to Assyrian Dictionary

2016-12-04T11:40:00.003-08:00

Goal: to create a modern high quality English to Assyrian dictionary, serving both Eastern and Western Dialects

Product: http://sargonsays.com

Motivation:

There are a couple Assyrian dictionaries that let you type in an English word and get the Assyrian equivalent. They’re uber useful but I wanted to address and nail down the following:

No site presents both the Western and Eastern Aramaic dialects
Search quality and relevance. I found just 1 site that gives really good search results. The other search result sites had to be paginated and manually searched through which is a big pain point for users
The main source of data comes from assyrianlanguages.org . Can't say enough about how good the data on this site is
Audio player, phonetic pronunciations, and definitions

Technologies:

View Layer: React.js (Less CSS)
Hosting: Microsoft Azure
Backend app: parse platform with custom Express.js cloud code
Database: mongodb
- Index generated using Java with Lucene
Data scraping: scrapy (python)
Native mobile: Android and iOS development kits

How:

Step 1: Compile a list of the most commonly used english words

Step 2: Using python scrapy, scrape various Assyrian websites for retrieving translations, audio, and phonetic pronunciations of both dialects

Step 3: Create a ranking system for the results

Step 4: Present the results in a clean and simple web app

Step 5: Integrate a native experience on the Assyrian android app using android RecyclerView

Improvements:

machine learning: Taking a machine learning course. I'm excited about using learning methods to improve the classifiers to create even higher quality search results!
Manually curate the words to add more words into the dictionary.
admin portal to give language experts the tool to make this the highest quality English to Assyrian dictionary

Zephyr or TestRail for Test Case Management

2015-09-16T10:14:00.003-07:00

We have a couple hundred test cases. They're managed in a well formatted Google Sheet and it's time to upgrade to a test case management and execution system. Any thoughts on a lightweight and simple tool to use?

Things to think about:

JIRA integration isn't so important at the moment as we're not using JIRA but it could be important should we decide to move to JIRA (wouldn't be for another 6 months atleast)
Executing test cases from a mobile device is a nice to have
A configurable dashboard that shows the current health of testing activities across various apps and builds is ideal

Any thoughts on Zephyr or TestRail ? I'd love to hear your experiences with these tools

the all new Assyrian App for iPhone

2014-12-03T14:22:00.000-08:00

The time has come. For well over a couple months, I've been coding away a brand new version of the Assyrian App for iPhone. Users voiced their opinion about the original app, which has garnered near 50,000 downloads. And so we responded. We rewrote the entire app from scratch. And it's something we're very proud of and really hope that users will enjoy it as much as we do.

What's new? Well... everything

private messaging with notifications
group chats
profile photos
Phrases in both Eastern & Western Syriac dialects
bla bla bla (really, there's a ton of cool useful features in the app)

All the app's features can be viewed in this video.

Dear Apple, I plead you to please approve this app. Love you lots!

Thank you Three20. It's been great!

2014-11-08T21:36:00.002-08:00

Dear Three20 Framework & Developers (Jeff Verkoeyen)

Today marks the final day I'm using the Three20 iOS framework for one of our iPhone Apps (Assyrian)
Despite the numerous warnings provided on your archived repo github page years ago against using Three20, I went ahead and used it anyways. I used it because it got the job done.

It. got. the. job. done.

It provided iOS developers with enough boiler-plating to get up and running quickly with a fairly complete iPhone app. In fact, we 'templatized' an app generator script that would generate new Three20 based iPhone apps, complete with custom URL mappings. It all worked just fine.

Until iOS8. And even in iOS8, most of the functionality works great! You guys did an awesome job and I just want to say thank you. I'm using NimbusKit now and it's super nice. The dependency architecture is flexible and simple. Thank you for maintaining NimbusKit, providing a migration guide, and just being active in the developer community.

I'll toast to Three20 tonight and to the solid developers who created it, then developed NimbusKit, and beyond.