Micah Alpern

Susan Kare, you’re my hero. ♥️

@rohanpaul_ai My favorite "reverse prompting" I ask @meetgranola after a meeting, "What was the key truth that was left unsaid in this meeting." Shockingly useful insights.

Coinbase CEO, Brian Armstrong: Some great insights on how they are using internally hosted AI Agents. "It’s connected to every Slack message, every Google Doc, and every Salesforce data confluence. Now, this is all linked up and the data is all aggregated, so you can ask these agents questions. Every team is using it—legal, finance, everything. It’s like the "Oracle of Coinbase." I’ve started to ask it things that go beyond just simple prompting, like "Hey, can you write this kind of memo for me?" I’m asking these AI agents now, as CEO, "What should I be aware of in the company that I might not be aware of?" It will tell me, "Did you know that there’s actually disagreement on this team about the strategy?" I realized I didn't know that, but the AI does because it can read every Slack message and every Google Doc. Tobi, who is on my board, calls this "reverse prompting." Instead of telling the AI agent what you want to do, you ask it what you should be thinking more about." --- From @theallinpod YT channel (link in comment)

Is there a way to connect @claudeai to both your work and home email/calendar/linear? cc: @trq212

Google just shipped DESIGN.md — a portable, agent-readable design system file. That's the real announcement. Everyone's covering "vibe design" and the canvas. But Stitch now has an MCP server that connects directly to Claude Code, Cursor, and Gemini CLI. Your coding agent can read your design system while it builds. Google already shipped official Claude Code skills for this. The pipeline works today. A PM describes the business objective. Stitch generates the UI. The coding agent reads DESIGN.md and builds against it. No Figma export. No spec document. No "the developer interpreted the design wrong." PRD → design → code used to be three teams and three handoffs. Now it's one loop with one context file.

The cost of not tinkering and exploring has never been higher.

Photoshops new AI rotate tool works perfect!

it’s kinda crazy ppl don’t realize everything you ever encounter is a bundle of tradeoffs. products, people, etc. you buy into the whole package or you don’t. there’s no unbundled version waiting somewhere with the rough edges filed off. e.g. the reason apple products feel inevitable is that the constraints are actually the things that are load bearing. the thinness of the thing costs you the port or the battery life. the walled garden costs you freedom but buys you coherence. you’re not meant to want everything. you’re meant to recognize yourself in a particular set of sacrifices. the same logic applies everywhere.

@ClassicII_MrMac Is it hardware locked? Could you hack/upgrade components?

WHAT!?!?!?!? MacBook Neo Repairability: 1. The battery is fully secured by screws with no adhesive 2. Modular components USB‑C ports, speakers, and headphone jack 3. The keyboard is replaceable as an individual part rather than a full top‑case unit Is this the same company?

the human role in software keeps moving up the stack. the question is: how far up can we actually go? lukew.com/ff/entry.asp?2…

in just a few years, developers went from writing code to steering groups of autonomous agents that write, test, and ship it. the rest of us are next. what does that mean for the role of humans in AI products?

@__omura__ @akchunks でも、箱とパッケージがとても良いですね。遊舎工房のお店をこれからもチェックしておきます。私はカリフォルニアに住んでいますが、日本国内で手作りキーボードを購入できるように東京で私書箱を開設しました。

@malpern @akchunks おさかなキーボードの完成品は遊舎工房に不定期に入荷されるのですが、入荷数が毎回少ないです。 x.com/yushakobo_shop… 大西氏（x.com/IlllIlllIlIlIll）はオープンソースでおさかなキーボードのデータを公開しているので、自作する人もいます。 github.com/TakumaOnishi/F…

ZMK用のタッチディスプレイドングルを作りました * 画面上のボタンをタッチしてZMKのbehaviorを実行できます * Bluetoothの接続管理もタッチ操作できます * 時計としても動作します ドングル兼ディスプレイ兼マクロパッドとして使えます #自作キーボード

@__omura__ @akchunks なんてことだ、絶対に一つ欲しい！でも遊舎工房（Yushakobo）に行ってみたら、「魚キーボード」は在庫がないみたい。たぶん前のグループバイの商品だったのかな。どうやったら手に入るんだろう！🥹

@malpern @akchunks Osakana keyboard by Takuma Ohnishi o24.works/fish/

@CharlesPattson @DannPetty youtube.com/shorts/3Mx9XPC…

@CharlesPattson @DannPetty I still am hoping for an IRL design workshop/small conference for people to share and learn these new tools, mental models, & workflows together.

A new world. This is a peak into the future of building. It’s not finessing pixels in a tool, it’s live and direct — shipping straight to prod.

The Claude frontend skill outputs are already recognizable. Off-the-shelf aesthetics become cringe so fast. See Bootstrap, Material, etc etc. There’s no way around learning what goes into good design. That means open eyes, lots of practice. Easier than ever but fundamentals same

I like to imagine this is what grepping for something in a codebase would feel like to Claude

@michalmalewicz I still wish Apple had released a new 12 inch Air

I'm a professional. I use a Macbook Air (M4/32GB)

how to get started in electronics (without a degree) electronics = build → break → measure → fix 1. fundamentals → voltage, current, resistance ohm’s law power (v × i) series vs parallel don’t memorize → wire it → measure it → understand it 2. basic tools → breadboard resistors, capacitors, leds multimeter bench supply oscilloscope later 3. primitives → led blink voltage divider button + pull-up transistor as switch op-amp basics complex boards = combinations of these 4. microcontrollers → Arduino Uno or Raspberry Pi Pico learn gpio → pwm → adc → i2c/spi/uart hardware + code = leverage 5. schematics → read them → trace power → understand ground 6. debugging → measure first check ground verify rails isolate subsystems 7. pcb design → learn KiCad design → order → solder → iterate you will burn parts → misread datasheets → short rails that is the training.

presenting my new talk, Designing AI Applications on lessons learned building AI-native products and companies over the past 4 years, for free on Wednesday March 5th.

@mass_0X00 どうやって注文すればいいですか？

偶然ですが誕生月なので嬉しい☺️ #キーボードのカレンダーを作ろう2026

The design orgs that will thrive in 2026 and beyond will have: 1. Execs leading from front with tools, experimenting, into the weeds with general capabilities. Not fear mongering or stressed about "uncanny valley" effects of the tools. 2. Leadership creating protected time to learn regularly for ICs. 3. Bigger education budget spend for tooling, courses, learnings. 4. Actively disengaged with static mocks/prototypes -> fully bought into the real thing. 5. An AI "counsel" that helps push strat forward.