I love this! data nerds, pls enjoy

Genuinely very impressed by the SVG of a pelican riding a bicycle I just got out of Google's new Gemini 3 Deep Think model simonwillison.net/2026/Feb/12/...

I had the hypothesis that harness is (as or) more important than the model, but couldn't prove it. Here is the proof! An excellent article. Must read. > The model is the moat. The harness is the bridge. Burning bridges just means fewer people bother to cross. blog.can.ac/2026/02/12/t...

We are still there in the Show HN main page - after 9 hrs of posting! 🔥 cc @georgeck.me ! news.ycombinator.com/show

We launched HN Companion hncompanion.com, the standalone version of the browser extension that @georgeck.me and I built a year ago 🚀. Read HN better with Vim keys and HN-aware summaries that understand thread hierarchy and up/down votes. Here is a sample post: app.hncompanion.com/item?id=4693...

Here is a spec that follows the suggestions in the post. Walk the talk, right? github.com/annjose/agen...

Usually we say... Trust, but verify Now with coding agents, we say... Steer, but verify

Wrote a post on what I learned from doing agentic coding. Not a playbook. Not "best practices". Just what's helping me build things right now - after shipping apps like HN Companion, Pomodoro Flow, Beast Mode workout tracker.🚀 Hope it helps you too. Learn👏Teach👏 Learn annjose.com/post/agent-c...

It has been a while since I wrote about agents, so here are my latest thoughts. I have a lot less time than I did eight months ago, so this is far more informal. crawshaw.io/blog/eight-m...

Published a new blog in the ATmosphere - pckt.blog/b/ann-learning. I want this to be casual than the main blog annjose.com. Share anything that I learn, whatever comes to mind. The first post: Run Claude and OpenCode with Qwen3-Coder pckt.blog/b/ann-learni...

You can do the same with OpenCode opencode.ai too. 1. Create ollama API key ollama.com/settings/keys 2. Launch OpenCode 3. Run `/connect`, choose Ollama Cloud provider 4. Choose any model, eg: qwen3-coder:480b 5. Code away! 🔥 It did well in basic coding. Want to try using this for smaller tasks.

TIL: You can.... run Claude Code ... with open weights model qwen3-coder-next ... using ollama ... on the cloud ... 🔥 This is how you do it. Create an account in ollama.com. Then do... ollama signin ollama pull qwen3-coder-next:cloud ollama launch claude --model qwen3-coder-next:cloud

TIL: `pnpm` has a `why` command that shows you the dependency chain of a transitive dependency. Eg: to see how @isaacs/brace-expansion ended up in my project, I do: pnpm why @isaacs/brace-expansion 2>&1 Output:🚀 devDependencies: wxt 0.20.13 └┬ minimatch 10.1.1 └── @isaacs/brace-expansion 5.0.0

Everyday I am more convinced that I should wait a bit after each wave of AI hype before I try to run with the wave. From "Developers are obsolete" to "OpenClaw is the perfect AI assistant and everyone should have it". The world has changed 🙌 1password.com/blog/from-m...

Agentic coding changed how I build. It took time to understand the concepts and how they fit together. Then my projects took off - esp. HN Companion hncompanion.com w/ @georgeck.me🚀 I wrote about these concepts and how to use them to build solid apps with quality code! annjose.com/post/agentic...

Great video - Why is knowledge getting so expensive? - by Jeffrey Edmunds, who worked as librarian in Penn State University for 35 years. An important concept explained in clear terms. I took notes at the end of the video and sharing it in the thread below... www.youtube.com/watch?v=PygU...

TIL: In 2013, Open Whisper Systems built [0] an encryption protocol to secure messages in their app TextSecure. In 2014, another app partnered with them to adopt it, completed [1] in 2016. First app became Signal. Second app? WhatsApp. [0] signal.org/blog/advance... [1] signal.org/blog/whatsap...

Oops, I started a new project: Skyreader, an RSS reader on the AT Protocol. Share cool articles like it's 2010 and Google Reader would never die. skyreader.app www.disnetdev.com/blog/2026-01...

Had fun building HNCompanion site using a 'council of agents' and git worktree. Prompt all agents to build the same feature, ask them to compare each other's solutions and pick the best. Review. Commit. Ship. 🚀 Speed of execution with this setup is unreal. There is no going back.🔥 hncompanion.com

We just released 𝚛𝚎𝚊𝚌𝚝-𝚋𝚎𝚜𝚝-𝚙𝚛𝚊𝚌𝚝𝚒𝚌𝚎𝚜, a repo for coding agents. React performance rules and evals to catch regressions, like accidental waterfalls and growing client bundles. How we collected them and how to install the skill ↓ vercel.com/blog/introd...

Coding with agents Claude/Codex/Gemini via CLI or Copilot is insanely productive and fun. But there’s a special joy in just chatting with Claude in the browser for pure brainstorming, without worry of burning tokens🔥. So I do that too, eg: what are the options to add automated tests to a project?

TFW you’re cranking out code with Claude Code and see this in the terminal: “You’ve hit your limit · resets 2am (America/Los_Angeles)” What it means: Go to bed. It’s time to sleep. 😢 What you do: Fire up the Gemini CLI and keep going. 😀

New on my blog: "Why Study CS? Thoughts on LLM-assisted software engineering" kmicinski.com/claude-code-...

Hilarious! AGI is almost here 😅!

AWS CEO Matt Garman says companies should not cut junior developer roles for 3 reasons: 1. junior devs are better with AI tools than senior devs 2. it doesn't make sense as a cost-saving move 3. removing junior devs breaks the talent pipeline I agree #1 and #2 www.finalroundai.com/blog/aws-ceo...

A great example of coding with AI properly - get to a starting point quickly, giving clear instructions, reviewing code thoroughly, comparing various approaches, keeping good code, deleting bad code mercilessly, optimizing for speed and test coverage, writing tools as needed & adding CI/CD 🚀 Super!💪

Fantastic interview with @kelseyhightower.com. Discusses DevOps, open source, AI hype and personal growth. "All paths lead to self-learning" "When people wear a title, they forget to evolve" "Senior engineer, junior human" The story about his first contribution to Kubernetes is inspiring 🎉!

Agreed. Definitely worth reading. It speaks directly to the question I have been asking the universe: instead of saying AI will replace humans, why not say AI will augment humans? Cory explains this through the concept of centaur vs reverse-centaur, and the biggest myth that AI can do your job.

If you're procrastinating upgrading your app's Next.js version to apply the RSC security patch, it's super easy. Just: 1. run npx fix-react2shell-next (refer link below) 2. update eslint-config-next to match Next.js version 3. run npm install Commit. Push. Deploy. 🚀 nextjs.org/blog/CVE-202...

I've been driving Zed as my daily IDE for about a week and while it has improved greatly since the last time I used it, it still has quite a lot of issues that make me less productive or, sometimes, even force me to reach out back to VS Code. Let's talk about those.

Great post by @chris.blue materializedview.io/p/mcp-server...

I remember the day watching the news in shock and disbelief. Just regular people who went to work. How do you even begin to make sense of that? My heart goes out to everyone who lost someone and to the first responders who saved lives. I can't imagine your pain, but I feel it with you today.❤️ #9/11

Happy birthday California 🎂🎉!

Nobody: Are you rooting for Alcaraz or Sinner? Me: I am rooting for 5 sets, that's it🔥 These guys are killing it! ❤️💪 #UsOpen

Bracket City's UI used to be simple and subtle. I used to love that. But the new UI is different and bad. Homepage with deep orange background with black text Puzzle page with caps lock keyboard, lower case answers and fluorescent blue highlights.😡 It takes away the joy of doing these everyday.😔

Felix, I believe in you! 💪💪 #USOpen 🔥

Not sure why, but lately Claude in browser is generating several versions of an answer instead of just one per request. For eg: I asked it to adjust my CLAUDE.MD file, and got 5 drafts at once 🤷‍♀️. This increases output tokens unnecessarily and in no time, it says you reached the 5-hour max limit 🙄.

Here is a typical chat with agentic coding for a big refactor (GitHub Copilot Claude Sonnet 4). I give multiple tasks in single shot 'coz each instruction eats up Copilot Premium request budget ($0.04/inst). And agents are capable of doing multiple complex tasks, creating a plan and executing it.🚀💯

ATProto Meetup in the Bay Area - happening right now. Good vibes! @rose.bsky.team is here! @knowtheory.net giving an intro to all apps built with ATProto With @georgeck.me lu.ma/mf4n7jbi?pk=...

I see that GitHub Copilot Agent explains its thinking as it codes - feels collaborative and transparent, like a good peer programmer.🎉 Claude Code is more proactive, often suggesting the next step - feels more autonomous.🚀 Both are great in their own way. Ideally, we want the best of both worlds!💪

Coding my workout app (Beast Mode) while Wimbledon’s on, waiting for Alcaraz in the Men’s Fourth Round. Shipped 2 features - Preview mode, Bowflex workouts. Bouncing b/w GitHub Copilot Agent and Claude Code. Love this dev + tennis combo! 💪🎾❤️ #wimbledon #vibecoding beast-mode-journal.vercel.app

Trying out Wimbledon's new AI chatbot 'Match Chat' 🎾(powered by IBM watsonx) - very handy for match stats and insights, answered most questions well. But for this specific match that ended in Thompson retiring due to injury, it struggled and kept asking about 'Match Context'🤷‍♀️. #wimbledon #context

Super happy to see this in my inbox today. >> Content Independence Day: no AI crawl without compensation! Thank you @cloudflare.social for doing the right thing! 👏 Credit where credit is due 💪.

Built a quick app to track workout loads/reps during sessions. Vibe coded using Claude Code and deployed in under an hour. Had to do a few dependency tweaks, but it’s a great start & ready for tomorrow’s session. 💪🚀 Building personal apps with AI is wild and fun! 🎉❤️ beast-mode-journal.vercel.app

A clear, thoughtful overview of recent LLM developments, compared side-by-side using a consistent benchmark and great insights on the risks of MCPs. Hard to believe how much has happened in just six months. You can watch the video or read the detailed transcript - highly recommended either way.💪

Launching a new blog series: Mobile On-Device AI 🚀 📱 While cloud computing is driving many AI breakthroughs, a significant opportunity is unfolding right in our hands - running AI locally on our ur mobile devices to faster, more private, and smarter apps. The main post: annjose.com/post/mobile-...

Walked 70 miles on the treadmill and 170 miles total (with outdoor walks) this May - an average of 5.5 miles a day! ❤️🎉 Long walks with country music outdoors and YouTube videos/TV shows on tread. Kept showing up. Small steps really do add up. 👟💪 Happy, grateful, proud. (Fitbit, Peloton stats📊👇)

The Golden Gate bridge in SF opened this day (27 May) in 1937 🌉 youtu.be/t_sm0OHQKrg?... A week long celebration of parades, fireworks and fun🎉 www.goldengate.org/bridge/histo... Learned from today's bracket.city, my favorite puzzle these days. (Sorry for the spoiler if you haven't finished yet.)

Remembering and honoring the fallen - those who gave their lives for this country and our freedom. Grateful today and always ❤️. Ceremony at Los Gatos Memorial Park.

Want to expand your knowledge base using #RAG, but but don’t want to deal with the infra overhead? Say hello to AutoRAG, a fully managed end-to-end RAG pipeline recently announced by Cloudflare @cloudflare.social I wrote a post on how it works (and how easy it is): 🔗 annjose.com/post/cloudfl...