Mutsumi

Software Engineer | Tokyo | Go, Ruby, Python, CI/CD, Product Engineering, Enabling, SRE | Planning to relocate to Germany/Netherlands/Singapore

[Mutsumi @mutsumi_dev]

Brief introduction:
A Backend-focused Software Engineer in Tokyo🗼🇯🇵

Working on modernizing our legacy Ruby/RoR system to a newer Go system, using LLMs in product features with Python/Celery worker, improving CI/CD pipelines, and structuring AI development workflows.

NTMU😊🤝🏻

[Phuong Le @func25]

I thought this article was pretty basic, but it's getting quite a lot of views: https://t.co/iTM5qs3r82. There might be some gold in it after all.

Originally wrote it as a foundation for more advanced articles, but maybe it stands well on its own?

[Jarred Sumner @jarredsumner]

Rewriting Bun in Rust

https://t.co/Rl8bcaxBFc

[sora🇪🇸エンジニア@Barcelona💻 @skyTrilingineer]

リポ��トしたら無料です！

スペイン地元企業への転職奮闘記🇪🇸 | sora💻エンジニア@バルセロナ🇪🇸 @skyTrilingineer #note https://t.co/LCg1XfapSZ

[Thariq is on vacation @trq212]

I’ve found the most important part of working with Fable is discovering my own unknowns so I can prompt it better, heres how I do that.

[Mutsumi @mutsumi_dev]

Welcome back!

[Claude @claudeai]

Fable 5 is back.

[Mutsumi @mutsumi_dev]

Finally ❤️‍🔥

[Thariq is on vacation @trq212]

excited to get Fable back in your hands!

[Gergely Orosz @GergelyOrosz]

It's live!

• YouTube: https://t.co/UY2BRCv5uY

• Spotify: https://t.co/0LMBXZNUoB

• Apple: https://t.co/0VYMmVcVzj

[Mutsumi @mutsumi_dev]

Actually, I applied to some companies and passed the screenings. I was surprised that they had tech interviews, especially Leetcode-style coding interviews, even in 2026!

However, as @neetcode1 mentioned in the video, the companies are in the process of rethinking.

[Mutsumi @mutsumi_dev]

Hey! This is the collaboration that I wanted to see!

I haven't watched it yet, but I will do it and am looking forward to knowing how they think about tech interviews!

Tech interviews with NeetCode https://t.co/9vhjcMY5eA via @YouTube @GergelyOrosz

[Mutsumi @mutsumi_dev]

Wow! Insane!
Cannot see it for now, but it's definitely what people want to get!

[Thariq is on vacation @trq212]

Slack now renders HTML attachments instead of just showing it as text 😭🙏

[Mutsumi @mutsumi_dev]

@bcherny Wonderful release on the day of Claude event in Tokyo 🗼!

[Mutsumi @mutsumi_dev]

@trq212 I wanted to join, but couldn't join my family event...

Hope you all have a wonderful day!

[Mutsumi @mutsumi_dev]

Wow! I didn't know that, and I'm going to jump into the red ocean...🫠

[The Pragmatic Engineer @Pragmatic_Eng]

Ever wondered why it's much easier to get a job January to June than July to December? Fresh data from our State of the software engineering job market in 2026 shows why: because historically 90%+ of hiring happens the first part of the year.

More: https://t.co/AjZenRUyj6

[Mutsumi @mutsumi_dev]

@bcherny How about the /goal?
If you use it, I'd like to know how you distinguish them.

[Mutsumi @mutsumi_dev]

I wonder whether there is any alternative to obra/superpowers.

The update has been stopped for 2 weeks. Does everybody just use simple Claude Code or Codex?

[Mutsumi @mutsumi_dev]

@jeremybernier Welcome to Japan!

Hope you have a wonderful time here.

I live in Tokyo 🗼, so if you have a chance let's have a coffee ☕

[Andrej Karpathy @karpathy]

This works really well btw, at the end of your query ask your LLM to "structure your response as HTML", then view the generated file in your browser. I've also had some success asking the LLM to present its output as slideshows, etc.

More generally, imo audio is the human-preferred input to AIs but vision (images/animations/video) is the preferred output from them. Around a ~third of our brains are a massively parallel processor dedicated to vision, it is the 10-lane superhighway of information into brain. As AI improves, I think we'll see a progression that takes advantage:

1) raw text (hard/effortful to read)
2) markdown (bold, italic, headings, tables, a bit easier on the eyes) <-- current default
3) HTML (still procedural with underlying code, but a lot more flexibility on the graphics, layout, even interactivity) <-- early but forming new good default
...4,5,6,...
n) interactive neural videos/simulations

Imo the extrapolation (though the technology doesn't exist just yet) ends in some kind of interactive videos generated directly by a diffusion neural net. Many open questions as to how exact/procedural "Software 1.0" artifacts (e.g. interactive simulations) may be woven together with neural artifacts (diffusion grids), but generally something in the direction of the recently viral https://t.co/z21CP5iQfu

There are also improvements necessary and pending at the input. Audio nor text nor video alone are not enough, e.g. I feel a need to point/gesture to things on the screen, similar to all the things you would do with a person physically next to you and your computer screen.

TLDR The input/output mind meld between humans and AIs is ongoing and there is a lot of work to do and significant progress to be made, way before jumping all the way into neuralink-esque BCIs and all that. For what's worth exploring at the current stage, hot tip try ask for HTML.

[Addy Osmani @addyosmani]

https://t.co/FSx0reCCLH

[Mutsumi @mutsumi_dev]

A New Episode has been released!

I personally, to be honest, didn't know much about Pi 😅

But, the following blog helped me to understand what Pi is
https://t.co/Ts8Wyb35vv.

Excited to watch the episode!

[Gergely Orosz @GergelyOrosz]

OpenClaw - the agentic software spreading like wildfire - was built on top of Pi, a minimalist, self-modifying agent. I sat down with Pi's creator, @badlogicgames and longtime Pi user (+ the creator of Flask) @mitsuhiko to talk Pi, and their (very grounded!) takes on building with AI.

Timestamps:

00:00 Intro
07:30 How Mario, Armin, and Peter Steinberger met
15:15 How 30 dev teams use AI agents: learnings
21:50 The importance of judgment
24:26 Challenges when non-engineers write code
28:30 Downsides of over-automation
32:18 Pi
48:09 OpenClaw + Pi
50:54 “Clankers”
57:32 Open source and AI
1:00:22 Complexity as the enemy
1:02:50 Building an AI-native startup
1:11:52 “Slow the F down”
1:16:40 MCPs vs. CLI
1:25:03 Predictions and staying up to date

• YouTube: https://t.co/u9n7ePTaAO

• Spotify: https://t.co/TvbqPnbfNz

• Apple: https://t.co/4ACETLJ1Zm

Brought to you by:

• @statsig  – ⁠ The unified platform for flags, analytics, experiments, and more. https://t.co/ZCSOIcWv31

• @SonarSource  — The makers of SonarQube, the industry standard for code verification and automated code review. Try it out for yourself. https://t.co/QtBhYDH9UX

• @WorkOS  – WorkOS gives you APIs to ship enterprise features – SSO, directory sync, RBAC, audit logs – in days, not months. Visit https://t.co/jhFNq3a7n7 to learn more.

---

Three parts I found especially interesting in this discussion:

1. New trend: AI makes it harder for senior engineers to reject pointless complexity.

Historically, senior engineers kept software complexity at bay simply by saying “no” a lot. But Armin observes that these days, more junior engineers and product managers deploy agent-scripted counterarguments when a senior colleague kicks an idea to the curb. This makes decision-making exhausting, and more bad ideas make it into production as a result.

2. It should be MUCH easier to build specialized tools for specific tasks.

Different projects need different harness types because, as Mario points out, the same hammer is not ideal for every single construction job. As such, Pi is built with the goal of allowing the creation of specialized harnesses. It can modify itself so that a user can create the bespoke harness needed for any task. Mario believes it’s a preview of how self-modifiable software might look in the future.

3. Automation bias is one of the biggest risks of working with AI agents.

Once devs confirm that an AI agent can produce acceptable code, they start to review its output less often, even though agents can – and do! – produce slop. Mario advises being far more sceptical with agents, and cautions that the quality of their output isn’t guaranteed, however well they performed previously.

[ClaudeDevs @ClaudeDevs]

Claude Code can now send push notifications to your phone when a long task finishes or Claude needs your input.

Walk away from the terminal, we'll let you know when it's done.