On a daily basis, my morning started with a cup of home-brewed coffee.

I have a notebook for tips, recipes, favorite beans, and how they are combined to make a good cup of joy.

Things were fine until we set up a brewing station in our workspace.

Sometimes we had a new bean, and I couldn’t remember the right recipe for it.

The notebook is at home. I need an app for it.

So I got Claude Code to plan the app with me, using the Kiro method.

For the app foundation, I used Augment Code. Its ability to index the codebase saved me many tokens, especially with a long task list.

I switched to Claude Code for fine-tuning the UI of each view. I also got my hands dirty learning Swift UI.

I named the app Taste because nothing beats that first taste of the day's first coffee sip.

Here’s how it works for me. I test every new recipe in a brew session. Each session gets saved to my Brew History with a note. Once I master a recipe, I move it to my Recipe list. Then, both my history and my saved recipes are ready for the next brew.

It cost me two weekends to develop the app that became my morning routine's sidekick from then on.

I have learned a lot from the development.

More importantly, I can now leverage AI as my BA, Developers, and Testers to build anything I want. Whether a prototype, an MVP, or a production-ready app, they are proof that “vibe-coding” is a myth. The idea that an AI will build something for you from a single prompt is false. From my experience, Context Engineering matters. Product Requirement Prompt engineering matters. And mastering the AI development process matters.

It's always best to learn when you start building something for yourself.

Woke up this morning to the news of ChatGPT Atlas's landing. As a product guy, my first thought was: classic product-led growth. The strategy quickly pivots to user retention, as I discovered Atlas is exclusively for Plus, Pro, and Business users. Smart move.

This immediately brought to mind my daily driver, Perplexity Comet. I use it not just because it's an AI browser, but because it's a *neat* AI browser. Built on Chromium, it inherits all the fundamentals but with a clean, seamless design. When it comes to AI-assisted browsing, it feels like I'm "vibe-browsing" the same way I "vibe-code" with Cursor. My point is, you have 100% control over Comet, just like your coding IDE. It's a native experience.

ChatGPT Atlas, on the other hand, feels for now like a browser with a Copilot sidebar. Think of it as installing a Copilot plugin in Obsidian—it's an LLM answering questions about your main tab. Useful? Absolutely. But is it compelling enough for a browser switch, or even as a dedicated secondary browser? I'm not so sure.

The one major upside: no waiting list. Atlas is here now for Mac users. This stands in contrast to Claude for Chrome, which has yet to see a confirmed release in 2025 despite a long waitlist. This highlights the effectiveness of OpenAI's approach. They seize the market-mover advantage by prioritizing speed to market, whether the product is an MMP or a full-featured release.

And just like that, the AI Browser Wars are on. But let's be real, everyone is just waiting for Google. You can feel the pressure on their product team from here. They own the browser core, a captive user base of billions, and all the AI advantages. When they finally ship, it won't be an MMP like Atlas; it will be a fully integrated beast. Success or not, their move will feel less like a new battle and more like a checkmate on the whole damn war.

Back then, creating those breakthrough moments that hook users was one of the key parts of every product I built or rebuilt.

Lately, I found it quite challenging when developing the user experience for AI Agents. The customer journey of AI Agents is short. From the cold start to the moment of trust, the AI Agents have only one chance to show off their "skillsets." Midjourney generates artist-like artworks. ChatGPT solves problems like an expert. Sonar Pro conducts high-quality research with fact-checked citations. Unlike the traditional products that can be optimized throughout their funnel, the AI Agent experience doesn't have this standout capability. Once a user message, which is considered the beginning of the funnel, is received, the response, no matter how satisfying, is considered the end of the funnel.

And then I noticed that the answer was already there: It's the moment of trust when AI Agents amazed the users.

Or, I can say: "When an aha moment becomes a trust moment".

Traditional products' aha moment can transition the user from step to step. At some touchpoints, the users found out that their needs are being fulfilled amazingly.

AI Agents shift that needs fulfilling from a transactional to a relational "Aha...".

There are three levels of trust:

- The "Aha" of competence: The users changed their impression that the AI Agents can actually help them. The games were levelled. The weak became stronger with a useful sidekick. Once they trust, they will return.

- The "Aha" of insight: This is the moment the AI delivers unexpected, proactive value. It moves from answering questions to providing genuine insight.

- The "Aha" of companionship: There are key insights that state Maxiaocia's users trust the AI more than their family. When it comes to money, having a fellow who doesn't judge the stupid question, who patiently explains about hard stuff, and who knows all your secrets but never gossips is rare.

As AI Agent builders, the main job is shifting. The job is not only about designing features, but also initiating these moments of trust that will turn a tool into a true partner.

Gen AI is transforming product development, revealing limitations in traditional methods, particularly in Customer Insight, UX Research, Business Analysis, Product Design, and Product Management. Workflows and deliverables can now be simulated and produced within minutes, significantly reducing the time from idea discovery to documentation. As a result, legacy gatekeeping processes are struggling to keep up.

At MoMo, senior management is empowering Product Managers by aligning gatekeeping processes with key business metrics such as revenue growth, customer retention, operational efficiency, and time-to-market. This approach enables the Product team to focus on delivering measurable business impact rather than simply completing initiatives.

Looking ahead, the future extends beyond rapid prototyping. The next phase is context engineering, which enables seamless handoff of MVPs across scrum teams. With AI tools generating product specifications, Product Managers can concentrate on high-level initiatives that balance user experience with business objectives.

If you are not yet leveraging AI to streamline your product management processes or quickly develop new product features, this may present a significant challenge. View AI as an innovative tool to enhance your workflow and ensure it works to your advantage. Even with these capabilities, this is only the beginning, as many developers are now using AI to further elevate their product development efforts.

However, building a proof-of-concept, whether a prototype or MVP, is only part of the process. Product Managers play an essential role in aligning business plans, product positioning, and operational, support, legal, and financial considerations—responsibilities that extend beyond what any prototype can demonstrate.

The role has evolved. It’s time our processes and mindsets do too.

Everyone is talking about GPT-5 models these days, calling it a "Claude killer." It seems that technical influencers are some of the most aggressive narrators in the industry. Whenever a new release is discussed, the features highlighted should ideally be game-changers. Remember the hype surrounding new social networks that were dubbed "Facebook killers" in the past? 🧐

After months of refining my AI development process, alongside Claude Code, I have established a procedure that maximizes the capabilities of the Anthropic AI-powered CLI tool. This not only offered vibe-coding but also empowered me, as a Product Manager, to build products similarly to real-world product development, which includes:

• Well-planned product requirements
• Clear process control
• Specialized execution management
• Consistent task coordination
• Quality assurance integration

I call it my 3-layer vibe-product-building approach:

First Layer: This involves defining a strict, repeatable workflow that standardizes the steps necessary for process compliance and consistent execution. Essentially, I instruct Claude Code to create the Product Requirement Document (PRD), generate the backlogs, define the development stack, write the technical documentation, and conduct grooming sessions. Sound familiar? Yes, I’m tired of waiting for Kiro, the vibe-coding IDE, so I borrowed the concept. Fortunately, it has proven to be very effective.

Second Layer: This layer leverages specialized AI agents, thanks to Claude Code’s sub-agents, for each execution of the workflow. It benefits from the specialized expertise, automated decision-making, and scalable capabilities of these sub-agents, which can develop new sub-agents based on the specifications from the first layer. Imagine having a team of experts join your scrum team for just $20 each, hiring the right one for every task—all made possible through the magic of AI.

Third Layer: Here, I focus on sprint planning for direct task execution from structured artifacts. Again, I apply a pure Kiro methodology without adding external complexity. There’s no need for Claude Flow or Task Master; Claude Code sub-agents tackle one task at a time and utilize Claude Code’s native to-do feature. This ensures clear deliverables, visible progress tracking, and high-quality integration.

A powerful AI model cannot deliver the benefits of scaling across teams and projects, maintaining quality without sacrificing speed, reducing decision fatigue through automation, or ensuring knowledge transfer through effective documentation. Not Claude, Gemini, or GPT can achieve this. A well-designed setup can.

To date, there is no AI-powered CLI tool that can set things up as effectively as Claude Code. So, welcome aboard GPT-5; let’s meet in Claude Code. I hope the Claude Code Router serves you well.

Last week, MoMo Travel’s Train Ticket Booking got a full upgrade — a backend refactor, a fresh new UX, and a 1 percentage point lift in conversion rate. Doesn’t sound like much? In a high-performing product, that’s a huge leap.

We didn’t completely reinvent the product, but we challenged ourselves with a tough question: “If this product weren’t ours, how would we improve it?” This mindset led to numerous small enhancements. When users recognized these improvements during testing sessions, it made all our efforts worthwhile.

We also didn’t wait for a big green light. Train Booking had always been “doing fine,” so full resources were hard to come by. Instead, we broke it down and tackled it one baby step at a time:

• Trình rewrote the wrapper for a cleaner, scalable codebase
• Nam rebuilt the entire promotion logic
• Ms. Hương, our talented Product Designer, crafted two prototypes with distinct user flows and selected the most valuable enhancements from them for the final shipment.
• Mr An, our awesome Tech lead, fine-tuned the discovery logic for a smoother entry touch-point
• Hòa and Trường didn’t wait for tickets — they shipped
• Nguyên patiently polished the ticket's QR visibility
• Bảo and Hoàng mapped error codes and rebuilt dashboards
• Oanh & Như, our calm QCs, hunted bugs with ruthless precision

We worked like a dream team — because everyone brought something different to the table:

• FE owns the user’s visual experience
• BE knows how the data connects to that experience
• QC sees the product from every user’s perspective

As long as they’re contributing, my job as a PO is to listen, align, and help us execute.

And it wasn’t just the product team. Our Business team joined grooming early, tracked the project’s movement, and worked with partners so we could stay focused. That’s what cross-functional collaboration looks like.

This wasn't my first remarkable project at MoMo, but I’m grateful to be involved. It's always a blessing to work with this incredible team, utilizing the frameworks and tools that support us, and being elevated by a culture where collaboration is embraced as the norm.

Today, I deleted two of my Linodes. Back then, I created one for my PHP-based projects and the other for my NodeJS-based ones. Since my day job is not development, I rarely log into any of them, as long as they are up.

I created a newly set up Nanode for n8n about three months ago. Self-hosting an n8n means you have to get your hands dirty with the bash commands again. I write .sh files for auto-laundrying the tiny Nanode to minimize the redundancy of disk sizes, docker resources, RAM... The scripts work well, and I wonder if I could do so with my long-time-no-see Linodes.

I spent all this weekend analyzing all the projects on the two Linodes. Firstly, I replaced most of them with n8n workflows. Then I migrated sites like this website into Docker containers, each one isolated within its sandbox, and the migration time was surprisingly and magically quicker than I thought.

After that, I kissed the no-longer-needed Linodes goodbye. Not only for saving a lot of monthly costs, but also for a neat working routine which is quick, centralized, and lean. This marked the day I stopped building things the traditional way as a developer, but instead, I began flowgramming them all as a flowgrammer.

Managed to be a creator again is one of the best things of 2025 for me. And, don't you ask, I am still sketching every day. Yay!

So you want to scrape a website, take a screenshot, or download an image. Well, cool kids nowadays are using n8n and Firecrawl for the same purpose. But there is a way cheaper solution, with Puppeteer. Puppeteer is cheap; all you need is a self-hosted n8n and a free mind to get your hands dirty a little bit, as below.

OS

Ubuntu 24.04 LTS

Dockerfile

# Dựa trên image n8n chính thức
FROM n8nio/n8n:1.99.0

# Quyền root để cài dependencies
USER root

# Cài Chromium và thư viện bắt buộc để render headless
RUN apk add --no-cache \
  chromium \
  nss \
  freetype \
  harfbuzz \
  ca-certificates \
  ttf-freefont \
  nodejs \
  npm \
  udev \
  bash

# Cài các thư viện NodeJS phục vụ crawl
RUN npm install -g \
  puppeteer-core@latest \
  axios \
  cheerio

# Khai báo path Chromium cho Puppeteer
ENV PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium-browser

# Trả quyền về user node
USER node

docker-compose.yml

version: '3.8'

services:
  postgres:
    image: postgres:15
    restart: always
    environment:
      POSTGRES_USER: ${DB_POSTGRESDB_USER}
      POSTGRES_PASSWORD: ${DB_POSTGRESDB_PASSWORD}
      POSTGRES_DB: ${DB_POSTGRESDB_DATABASE}
    volumes:
      - ./postgres_data:/var/lib/postgresql/data

  n8n:
    image: n8n-with-modules
    restart: always
    ports:
      - "5678:5678"
    environment:
      N8N_BASIC_AUTH_ACTIVE: ${N8N_BASIC_AUTH_ACTIVE}
      N8N_BASIC_AUTH_USER: ${N8N_BASIC_AUTH_USER}
      N8N_BASIC_AUTH_PASSWORD: ${N8N_BASIC_AUTH_PASSWORD}

      GENERIC_TIMEZONE: ${GENERIC_TIMEZONE}

      DB_TYPE: ${DB_TYPE}
      DB_POSTGRESDB_HOST: ${DB_POSTGRESDB_HOST}
      DB_POSTGRESDB_PORT: ${DB_POSTGRESDB_PORT}
      DB_POSTGRESDB_DATABASE: ${DB_POSTGRESDB_DATABASE}
      DB_POSTGRESDB_USER: ${DB_POSTGRESDB_USER}
      DB_POSTGRESDB_PASSWORD: ${DB_POSTGRESDB_PASSWORD}

      N8N_PORT: ${N8N_PORT}
      N8N_HOST: ${N8N_HOST}
      N8N_PROTOCOL: ${N8N_PROTOCOL}
      WEBHOOK_URL: ${WEBHOOK_URL}

      NODE_FUNCTION_ALLOW_BUILTIN: '*'
      NODE_FUNCTION_ALLOW_EXTERNAL: cheerio,axios,puppeteer-core
    volumes:
      - ./n8n_data:/home/node/.n8n
      - ./scripts:/data/scripts
    depends_on:
      - postgres

crawl-an-url.js

root@localhost:~/puppeteer-scripts# vi scripts/crawl-an-url-with-undetected-puppeteer.js 

const puppeteer = require('undetected-puppeteer');

(async () => {
  const inputUrl = process.argv[2];
  if (!inputUrl) {
    console.error('❌ No URL provided!');
    process.exit(1);
  }

  const browser = await puppeteer.launch({
    headless: 'new',
    args: [
      '--no-sandbox',
      '--disable-setuid-sandbox',
      '--window-size=1920,1080',
    ],
  });

  const page = await browser.newPage();

  try {
    await page.goto(inputUrl, { waitUntil: 'domcontentloaded', timeout: 60000 });

    // 👇 Dùng Promise-based timeout để chắc chắn tương thích
    await new Promise(resolve => setTimeout(resolve, 5000));

    const html = await page.content();
    console.log(html);

  } catch (error) {
    console.error(❌ Failed to load page: ${error.message});
  } finally {
    await browser.close();
  }
})();

n8n's Execute Command node

# Paste into Command input
node /data/scripts/crawl-an-url.js {{ $json.target_url }}

Note

Running n8n in the same place as Puppeteer is only for short-term usage; for long-term production, separate them.

There was a subtle comparison running through my mind while watching today’s Super Sunday match — right after that thrilling showdown between Everton and West Ham. Bowen scored a stunning goal. It wasn’t enough to save West Ham from defeat, but it was more than enough to remind me why the Premier League is so special. Beautiful football doesn’t just happen in clashes between the giants. It can come from any fixture, any team.

You asked ChatGPT about product-led growth (PLG) and, of course, it pointed you to the usual SaaS success stories — Slack, Notion, Figma. Sound familiar? Same here. I’ve worked at non-SaaS companies, so those case studies often felt... distant.
But here’s the thing: PLG isn’t exclusive to SaaS. At its core, PLG is about leveraging product-qualified leads — users already experiencing your product — to drive growth. Unlike sales- or marketing-qualified leads, these users are already in the door. Your job? Convert usage into revenue.

As a devoted Manchester United fan, I’ve spent more than my fair share of weekends buried deep in Premier League fixtures — waiting for some magic moment to get out of the cave. But beyond the emotional rollercoaster, I’ve also come to appreciate the Premier League for something else entirely: Fantasy Football. It's one of the most compelling, real-world examples of product-led growth in action — a role model in organic user acquisition, retention, and virality.
As one of the world’s premier sports properties, the Premier League capitalizes on marquee matchups — the big-name clashes every Super Sunday. But what about the lower-profile fixtures? How do you get fans to care about a game between Crystal Palace and Brighton? Marketing might bring in a few Son or Mitoma followers from Korea or Japan to the stadium — but Fantasy Football does the heavy lifting. It gives every match a stake, every player a storyline, and every user a reason to tune in.

First things first: Fantasy Football is a game built on real-world Premier League action. Players are given a virtual budget to assemble a squad using real footballers, priced according to their actual performance data. Each game week, managers make transfers, adjust lineups, and select their starting eleven — then earn points based on how those players perform in the actual matches. Bonus points come into play for smart choices like captain and vice-captain picks. The game even introduces small bets or rewards to raise the stakes. It’s strategy meets sport, powered entirely by real events.

I was hooked on Fantasy Football the moment the first game week results dropped. That unexpected thrill — like watching Danny Welbeck rack up triple the points of Erling Haaland — had me hooked. I started studying players more seriously. The deeper I researched, the more matches I watched. I dropped my bias for big names and discovered that players like Eze, Mateta, Wissa, and Elanga could deliver just as much impact as the league’s stars.

More importantly, I started treating every match equally. Sure, Super Sunday is still appointment viewing — but so is the final home game of Jamie Vardy at Leicester. Fantasy didn’t just deepen my engagement. It democratized it. Every fixture became a story worth watching.

One day, during a mid-game review on K+, the commentators casually mentioned Fantasy Football — and suddenly, everything clicked. It answered the question I’d always wondered: How do today’s commentators know so much about Premier League players, even those from mid-table clubs? Simple. They’re playing Fantasy Football too. Just like me. It also explained why their depth of knowledge doesn’t always carry over to leagues like Serie A or La Liga — because, like many of us, they’re fully immersed in the Premier League ecosystem that Fantasy creates.
From a product manager’s lens, Fantasy Football is a blueprint for product-led growth done right. No sales team, no pushy onboarding — users discover it, play it, and naturally pull others in. The product itself is the funnel. Its built-in virality kicks in through office leagues, group chats, and weekend banter — the more friends join, the stickier it gets. Engagement runs deep: weekly squad tweaks, transfers, captain dilemmas — all designed to bring users back, every single game week. And then there’s the emotional layer. This isn’t just gameplay — it’s pride, bragging rights, heartbreak. Users are invested. Add to that the network effect — where the product becomes exponentially better with friends in the loop — and you’ve got a living, breathing example of how great product design can drive growth at scale, with zero paid acquisition.

Sales and marketing teams often operate above the surface — focused on the visible tip of the iceberg, where results need to move fast. But when done right, a great product doesn't chase users — it compels them to play, return, care, and advocate. Fantasy Football proves this brilliantly. It shows us that some of the most powerful product-led engines aren’t flashy or loud — they’re quietly embedded in team colors, mid-table clashes, and last-minute transfers — hidden in plain sight, powered not by hype, but by product gravity.

As a Product Manager, from my point of view, art supplies are both the utilities for drawing and the products that went through their product development processes and successfully achieved the product-market fit. Since the day I started drawing again, I have done a bunch of research for the best art supplies. I found a lot of interesting kinds of stuff that are worth sharing, therefore this note.

The approach

My approach to pursuing art is to go with art supplies that are lightweight, reusable, and not too fancy.

• My art supplies have to be lightweight because I have serious backbone pain. Whether for sketching or watercolor painting, they should be portable and traveling-optimized. I like to draw everywhere, art supplies are one of the essential stuff in my go-bag.
• Stuff with reusability is always top of my criteria. I won't say it's because I care for the environment. I just try my best to not trash any waste that is a one-time use. Like drafting pencils instead of wooden ones (except Backwing's). Like fountain pens instead of ballpoint pens (except the Pentel Super Multi-8). I decided to buy Blackwing pencils or add the Pentel Super Multi-8 to my wishlist, mostly because of their production design which I admire the most. I think it's the blindside of everyone who does product development in the world. Yeah, I can hear you laugh, shame on me haha.
• It wouldn't be a research if I just fancy buying the top options of each category. Not only doing it will cost me a fortune, but also I love the "aha moment" when I figure out the hidden gems. I bought my Tombow Mono Graph almost ten years ago. I just found out that it is my only drafting pencil that can protect the lead yesterday. We will get back to this story later.

The categories

To create good art, I will need to improve my sketching, inking, and watercolor skills. So, the categories I am looking at are:

1. Sketchbooks & papers
2. Pencils
3. Pencil sharpeners
4. Fountain pens
5. Multi-liner pens
6. Watercolor brushes
7. Watercolor paints
8. Other watercolor supplies
9. Miscellaneous

As the title of the note already revealed, the first picked category is Pencil sharpeners. I am forty-something now but I still remember the first time I was using a sharpener on my pencil. Before buying this sharpener for me, my Dad still sharpened my pencil with his knife. A pencil sharpener was a precious thing back then, to everyone. The magical moment of having my pencil gently sharpened with a straight angle, from the wooden part to the graphite's lead point, would never fade out of my mind. In honor of that good memory, let's get started.

What do I need from a Pencil sharpener?

There are three factors when considering a sharpener:

• The supported range of pencil sizes: There are 2.0mm - 5.6mm leads, alongside a wide range of up to 10.5mm wooden pencils. I only need the sharpeners for two sizes: the traditional 8mm wooden pencils and the 2mm leads
• The offered features: The support range of the pencil shapes: hexagon, triangle, and circle. The sharpening angle. The gentleness of the shaving shows the quality of the blades. And the sharpness of the lead point. I only care about the sharpness of the lead point, to me, it too reflects the quality of the blades.
• The compartment of pencil shavings and dust. I need my sharpeners to go with a compartment for my traveling purposes.

The initial pencil sharpener line-up

I call them the initial line-up because they were randomly bought when I had no idea about the pencil sharpener. The MUJI one is because it's a stuff from the MUJI brand. The Spide-man one should be a gift for my son. The Faber-Castell one was bought based on Mateusz's Art Supplies Box, as you can see, I bought the mini version of the suggested one. The Helix Oxford and the Stabilo ones were suggested by some sellers at random art supply stores. Almost sharpeners in my initial line-up surprisingly still survive until now for one reason: They are somewhat working well. Except for the Stabilo Exam Grade 2mm Lead Sharpener, since the sharpener is made for students to use in examination sessions, it could give me the sharp long-point tip I need for drawing.

MUJI Manual Pencil Sharpener

The MUJI Manual Pencil Sharpener made it to be one of my favorite sharpeners. It produces everything I need for my pencils, the sharpening angle, the shaving gentleness, and the long-point tip. Using the sharpener is so easy and convenient: Just drag out the pen port, and plug the pencil into the shaving hole through the penport, then swing the handle. Your sharpening progress is done when the pen port is completely retracted to the sharpener body. All pencil shavings and dust will be retained in a big compartment.

Too bad, the MUJI Manual Pencil Sharpener is too big for a to-go sharpener, otherwise this note wouldn't be written.

Faber-Castell Sleeve Mini sharpening box

I decided to buy the Faber-Castell Sleeve Mini sharpening box after watching this video which you can check below of Mateusz. The suggested sharpener was one of his picks for his art supplies box. I searched on Shopee and found out it only cost me less than a dollar to buy it. I didn't know that the one I found was a different version of the Faber-Castell Faber 9000 sharpener, the sharpener that Mateusz suggested.

To me, the only difference between the two versions is the one I bought doesn't have the titanium-plated knives. This shouldn't be a big problem since it performed well with my HB pencils. I wish it could do the same with pencils with darker grades, which have more softer leads. It breaks the leads unpredictably. I wanted to know why, so I started my research to find the sharpener that fits me the most.

The research outcome

From my perspective, I don't believe anyone can overthrow the domination of the Germans.

• No one could beat them when it comes to blades
• No one could beat them in pencil sharpeners' product design
• No one could beat them since they were the veterans in the field

I stumbled upon it when I learned about KUM, the company that had a century-long production of pencil sharpeners, thanks to this video of Teoh Yi Chie. 7 years ago, Teoh made his statement that the KUM Automatic Long Point Pencil Sharpener was his best choice of portable pencil sharpener, and I couldn't agree more.

The two popular art suppliers Staedtler and Stabilo happen to be German, too. Staedtler was found in 1835, and Stabilo was found in 1855. Yeah, we are talking about companies with around two hundred years of experience in the industry. So, in "German steel blade" we trust!

Product design talk, did you see KUM The Masterpiece aka KUM Meisterwerk? The sharpener has two holes, hole number one is for removing the wood surrounding the lead, and hole number two is for sharpening the lead. The blue stopper gives you the indication to not oversharpen your pencil lead. The "aha moment" is when you figure out that you can remove the blue stopper to get the long lead, as long as you take all the responsibility afterward.

The best to-go sharpeners

As mentioned before, the to-go sharpeners that fit my needs should provide gentle sharpening, sharp long-point tip, and be able to retain the pencil shavings. Below are my picks after the research:

• For wooden pencil sharpening only: Blackwing One-Step Long Point Sharpener
• For both wooden pencils and 2.0mm leads: KUM Automatic Long Point Pencil Sharpener

Blackwing One-Step Long Point Sharpener

If Blue Bottle Coffee is the Apple of the coffee industry, then Blackwing is the Apple of the art supplies world. The company is well known for its top-notch pencils. I got some of those, but we will save the Blackwing pencils topic for other notes. Seeing the Blackwing One-Step Long Point Sharpener is like seeing the next iPhone to me. Its product design is neat. At the time, I can easily see that the sharpener fits my needs in every criterion. When I tried it the first time, I finally understood the term "German steel blade" completely.

From the name of my favorite Blackwing sharpener, you can easily guess that if there is a "one-step" sharpener, then it should be at least a "two-step" sharpener, right? They have it. The Blackwing Two-Step Long Point Sharpener offers the same feature just like the KUM The Masterpiece that I mentioned before, plus a compartment.

Even though the Two-Step one also has a lit product design, I decided to go with the One-Step. One step is quicker than two steps, right? That's right, I am good at Math.

KUM Automatic Long Point Pencil Sharpener

Using Blackwing pencils is my special exceptional allowance for me based on my favorite for stuff with damn good product design. I mostly use drafting pencils because they are environmentally friendly. I am using 2.0mm leads for sketching, and 0.3, 0.5, 0.7mm leads for detailing. Sketching with the 2.0mm led me to this one issue: I need to sharpen my 2.0mm leads.

I could easily find the 2.0 sharpeners. The most popular are Faber-Castell's, Ohto's, and the fancy Staedtler Mars Technico Lead Sharpener. The problem is the first two don't come with a "German steel blade", and the fancy Staedtler one is way too expensive, and it looks like it would suit the best with Staedtler 2.0mm lead holders. I have a cheap M&G Save the Tree 2.0mm lead holder and a clumsy MUJI 2.0 lead holder, I want my 2.0 sharpeners to work with them also.

Teoh Yi Chie enlightened me about the KUM brand and its amazing Automatic Long Point Pencil Sharpener. It's safe to say that the sharpener is the Blackwing Two-Step Long Point Sharpener plus the ability to sharpen 2.0mm and 3.2mm leads. My sharpener search reached an end after I bought the sharpener. However, I am now facing the very question of choosing which one every time I go out: The neat one or the full-feature one?

The takeaways

• The German offers straightforward features to keep a sharpener a sharpener. The Japanese make pencil sharpening funnier. Kutsuwa brand offers two blades in one hole for faster sharpening, the brand also allows you to dial the sharpener to adjust the sharpening angle.
• I got my jaw dropped when I found a lot of KUM's sharpeners in the Fahasa store on Shopee. They are all affordable for you if you wanna try. I can share with you the three options that caught my attention from my research: The Softie, the Cube, and the 4-in-1. Let's consider the first one is for an office tabletop, and the others are for a to-go.
• Everything has its museum. Here is a sharpeners museum in case you ask.

That's it for the pencil sharpeners. Almost 2000 words have been written. I hope you are reaching this line. Thanks for reading it. If you have any questions or wanna share your interest in pencil sharpeners or art supplies, write me an email at nguyenducban(at)me(dot)com.

Damn it, I thought I could survive writing the title, but I failed. So, long story short, we are moving back to Saigon by next January.

This is more a note-to-self post than a thought-sharing like others, but it's still worth a read, I guess 😋.

One night, I saw a video of a policeman and a cosplayer in Sun Wukong costume. I liked what I looked so I wanted to draw this.

The rough sketch which you can check below turned out not good. Yes, I am not an artist, so I don't expect the sketch to be as good as the artwork of the artists I am following on Instagram. I graded the sketch as bad because I couldn't either recognize the policeman or the Monkey King.

I sketched one hour a day every day before I got too busy to do so. I have studied other artists' styles and then somehow developed my chibi drawing style. Some of my friends are still using their chibi pictures that I drew as their avatars on Facebook.

Before I got my iPad Pro M1 and its companion Apple Pencil 2, I mostly sketched on sketchbooks, using pencils and colored pencils. I bought a lot of sketchbooks and art supplies, of course. Who didn't? 🧐 This behavior happened to everyone, I could say.

At the time when I figured out that my sketching skill was bad, alongside the fact that I found out that I had a bunch of idle art supplies, I also got inspired by some workshops that were organized for the artists who got involved in an upcoming art project in Hanoi. I got my chance to learn more about Dó paper, thanks to the Zó Project team. I came to the amazing exhibition of Vu Thai Binh to enjoy how the artist applied watercolor on Dó paper at his unique master level. I visited the lacquer studio of the HD team and the woodcut studio of artist Pham Khac Quang. After all, observing the progress of all the artists developing their pieces for the project initiated a sparkle in me.

I decided to get started with drawing again.

My approach is follow to the pathway that other artists stepped on:

• Start with the basis: Lines, rounded shapes, basic forms and structures, etc.
• Learn the hatching methods and apply them to the basic forms and structures.
• Study the proportions and anatomy.
• Practice, practice, practice...

From learning the basis, I found out that I picked up two of the top bad habits that should avoid to be better:

• I don't draw straight lines, I connect small lines.
• I overuse chaotic searching lines.

I found these very good tips here from Keezoo:

And below are the outcomes:

I also learned a lot about proportion from Keezoo, the man is not only sharing the basic tips, but also his practical tips that make life easier. One of the best was his "eye-level to ground-level as the base unit" tip. Gold! ✨

Below are some of the rough sketches which are much better than the rough sketch I shared at the beginning of this note. I know they are not the best, just want to share the progress (after three weeks). I need more practice, I know. But it's a good thing that you know that you are on a good track.

Stuffs doodling:

One of the best "hand tutorials" on YouTube is from Marco Bucci:

And the so-called outcomes (don't blame me haha):

So, from the rough sketch with all the short lines and chaotic search lines above, allow me to share with you the sketch that could be counted as my improvement, so far.

My target is to use watercolor to do urban sketching, and also draw some good quality pieces. The roadmap is still far, especially since my gap time is now close to its end. I hope I can manage to balance both working for a living and following the journey of making art in the future. I can't wait to share with you the incoming artwork I am now developing. I also want to share with you the way I tidied up my art supplies to have two sets for home drawing and to go drawing.

Thanks for reading.

Last week, my friend shared how his company switched from the full implementation of the Scrum framework to a much simpler implementation. My friend's company hired a professional project manager team, and they spent months learning how the company works before they suggested options. The framework showed its usefulness throughout the pandemic. After the lockdown order was lifted, it was still good for a while, but then some issues occurred:

• The company has four divisions across the globe in different continents which have working hours in different time zones.
• As a matured SaaS company, the company is now in the phase of operations and sales support. Of course, some initiatives require to building of new features on the roadmap, but they are too minor compared to others that are related to reacting to revenue-generation events. The Salesmen became the Business Owners who raised issues as Business Requirements. The Product Managers became the Technical Product Managers who worked directly with the developers for hot fixes rather than backlog those issues for formal Sprint sessions.
• High-level alignments are sometimes a myth that isn't based on the company product's core features, but from a PoC that only be proven by market research and then being sold after an impressive presentation with a Figma prototype. When it comes to development, it won't. The BOD already had some companies in their shopping cart. The next thing to do is to plan the implementation with new services from the newly joined teams.

Long story short, the Agile Methodologies are made for Product-Tech operations in order to plan, develop, and maintain products. There is no framework that fits a hundred percent with your entity. At the end of the day, only the wellness of the business operations matters. Any working process that fulfills this requirement would last.

Until new issues are found.

Back in the time when I worked as the Head of Product at YODY FASHION, I always reacted aggressively to anyone who came up with the idea of releasing the next yody.vn as an "MVP". I also banned the term internally of the Product team. Of course, I have my rationale.

The current yody.vn is a Sapo-based integration that carries all the experiences that the company needs for an e-commerce platform, respectfully. The only problem is that the Sapo-built platform doesn't offer what YODY needs, both short-term and long-term. Therefore, we didn't start with a product idea that needed to be proven for early adopters or beta users. We started with all the initiatives that matched our user needs, both the Operations team and the Customers.

An MVP is just for proving a product approach.

An MMP releases a Kraken into the ocean.

Of course, we will use a go-to-market plan to validate our current product-market fit, some phases of the plan could be misunderstood as the MVP, but they are NOT.

I told my dear product team that our goal is to release a product to the market, not to prove that our approach was right, or wrong. Therefore the "MVP" shit is off the table, my good fellas.

MMP, or we are out.

Sắp nghỉ VNG. Ủa?!! =))

Sau đúng 6 tháng quay lại VNG để làm-điều-mình-thích thì quyết định xin nghỉ. Nghỉ để tìm một chỗ mình có thể tiếp tục làm-điều-mình-thích. Đúng, sau 6 tháng vô cùng lean-in, đã thực sự ngấm, đã biết rõ rất là nhiều thứ. Và lần đầu tiên sau chừng ấy năm, lại muốn làm một điều gì đó ra hồn. VNG rất tốt, nhưng sẽ không làm được "chuyện đó" ra hồn.

Ngoài ra, 6 tháng qua cũng là khoảng thời gian đã bước ra khỏi vùng an toàn để đối thoại. Đối trong đối diện. Thoại trong hội thoại. Đã đối diện với rất nhiều người, đã nhìn thẳng vào mắt người ta bằng ánh mắt trong veo không ngại ngùng như trước đó đã từng. Đã giao tiếp tốt hơn với chừng ấy người bằng cách chọn ra những câu thoại phù hợp với thái độ hơn ngày xưa. Tốt hơn trong đối thoại không có giúp bản thân trở nên quan trọng hoặc hay ho hơn trong các mối quan hệ, nhưng nó giúp tin vào bản thân hơn, vì nó giúp thoải mái hơn trong cách giao tiếp, rành mạch hơn trong các cách cư xử và chỉn chu hơn trong lời ăn tiếng nói.

Sáng nay, Cường, đúng, vẫn là Cường, đã lại vào chat. Gần như lần nào chuẩn bị thay đổi công việc cũng sẽ nhận được message của Cường, bỗng dưng xem tử vi và thông báo về những tín hiệu khả quan. Có những chuyện không muốn tin cũng không được vì nó xảy ra quá nhiều lần và cũng đúng rất nhiều lần. Nếu lần này nó đúng, thì càng đúng cho cái nhiệt huyết thanh xuân đang rạo rực trong lòng.

Dũng giả thì không sợ khiêu chiến. Thắng thì đẹp. Thua thì tiêu sái.