Hacker News
A Round Up and Comparison of 10 Open-Weight LLM Releases in Spring 2026
Article URL: https://magazine.sebastianraschka.com/p/a-dream-of-spring-for-open-weight
Comments URL: https://news.ycombinator.com/item?id=47151455
Points: 1
# Comments: 0
Show HN: Seite static site generator with MCP server and Claude Code integration
I'm CTO at a startup, tired of maintaining five separate tools for our web presence while shipping software with Claude Code every day. So I built seite — a Rust SSG where the AI agent has a proper MCP interface to your site, not just file access.
seite mcp runs a Model Context Protocol server that exposes resources (docs, config, content, themes) and tools (build, create, search, apply theme) to any MCP-compatible agent. seite init also scaffolds a CLAUDE.md context file so the agent knows your schemas and conventions before it writes anything.
Everything else you'd expect: six bundled themes, collection presets for blog/docs/changelog/roadmap, one-command deploy to GitHub Pages/Cloudflare/Netlify, llms.txt and llms-full.txt on every build. Single binary, no runtime, sub-second builds. 331 tests.
MIT licensed, v0.1.6. Early but iterating fast — we're using it on our own startup's site so the feedback loop is tight.
curl -fsSL https://seite.sh/install.sh | sh Site: https://seite.sh Repo: https://github.com/seite-sh/seite
Comments URL: https://news.ycombinator.com/item?id=47151427
Points: 1
# Comments: 0
Mitochondria May Be the Key to Longevity
Article URL: https://www.nytimes.com/2026/02/19/well/mitochondria-longevity-health.html
Comments URL: https://news.ycombinator.com/item?id=47151414
Points: 2
# Comments: 1
Language Models will be Scaffolds
Article URL: https://alexzhang13.github.io/blog/2026/scaffold/
Comments URL: https://news.ycombinator.com/item?id=47151409
Points: 1
# Comments: 0
Raijin – coding agent harness written in Go [MIT license]
Article URL: https://github.com/francescoalemanno/raijin-mono
Comments URL: https://news.ycombinator.com/item?id=47151406
Points: 1
# Comments: 0
Show HN: First native zeroclaw build on Android/Termux (aarch64, no proot)
Zeroclaw is a Rust-based Nostr client/relay tool. As of a few hours ago, no one had compiled it natively on Android.
Gemini CLI tried. Gemini Android tried. Both failed.
The blockers: — koffi's build script passes `make -j` bare, which Termux's make rejects — The default linker OOM-kills during the final link step (Android blocks swapon for unprivileged processes)
The fix: mold linker + codegen-units=1 + lto=thin + opt-level=z in a .cargo/config.toml
Binary: 15.5MB. Build time: 23m 55s. Kernel: Linux 5.4.284-moto. Completed at 04:11:47 CST.
Repo with binary, build config, and full reproduction steps: https://github.com/BleakNarratives/zeroclaw-android
Comments URL: https://news.ycombinator.com/item?id=47151385
Points: 1
# Comments: 0
Electrical control of magnetism in 2D materials promises to advance spintronics
Article URL: https://phys.org/news/2026-02-electrical-magnetism-2d-materials-advance.html
Comments URL: https://news.ycombinator.com/item?id=47151384
Points: 1
# Comments: 0
A Curious Trig Identity
Article URL: https://www.johndcook.com/blog/2026/02/24/a-curious-trig-identity/
Comments URL: https://news.ycombinator.com/item?id=47151374
Points: 1
# Comments: 0
Trig of Inverse Trig
Article URL: https://www.johndcook.com/blog/2026/02/25/trig-of-inverse-trig/
Comments URL: https://news.ycombinator.com/item?id=47151373
Points: 1
# Comments: 0
Python Type Checker Comparison: Empty Container Inference
Article URL: https://pyrefly.org/blog/container-inference-comparison/
Comments URL: https://news.ycombinator.com/item?id=47151367
Points: 1
# Comments: 0
Jjq, a local merge queue for jj
Article URL: https://pauladamsmith.com/blog/2026/02/introducing-jjq-a-local-merge-queue-for-jj.html
Comments URL: https://news.ycombinator.com/item?id=47151364
Points: 1
# Comments: 0
Tech legend Stewart Brand on Musk, Bezos and his extraordinary life
The new Design for Stack Overflow is now live [beta]
Article URL: https://beta.stackoverflow.com/
Comments URL: https://news.ycombinator.com/item?id=47151351
Points: 2
# Comments: 1
Show HN: I scanned 35 SaaS products across ChatGPT, Claude, Perplexity, Gemini
I built a scoring system to measure how AI models represent software products when users ask buying questions. The process: I take a product, generate the queries a buyer would ask (category, competitor alternatives, head-to-head), run them through ChatGPT, Claude, Perplexity, and Gemini, then score how prominently the product appears in each response (0-10). Some findings from scanning 35 products: ChatGPT is the biggest blind spot. It scores 0 for most open source challengers, even ones with 30K+ GitHub stars. Incumbents dominate across all models. "Best project management tool" returns Jira, Linear, Asana — never Plane (31K stars on GitHub). Brand-name queries work. "Cal.com vs Calendly" scores 9+ everywhere. But generic category queries ("best scheduling tool") often return 0. Having revenue doesn't help. Trigger.dev raised $16M and scores 0/10 for background job queries. The scoring methodology: each model gets 0-10 per query based on mention position, detail, and recommendation strength. Product consensus = average across all models and queries. Code and methodology are open. Happy to scan any product if you drop it in the comments.
Comments URL: https://news.ycombinator.com/item?id=47151344
Points: 1
# Comments: 0
Illusion and Well-Being: A Perspective on Mental Health [pdf]
Article URL: https://faculty.washington.edu/jdb/articles/Illusion%20and%20Well-Being.pdf
Comments URL: https://news.ycombinator.com/item?id=47151341
Points: 1
# Comments: 1
UAA – A spec for AI and Human coding collaboration
I just made my own spec to solve vibecoding issues with modern apps.
This specification is made for structuring how AI agents and humans collaborate in software systems.
Well, I tried to vibecode but my codebase was growing messy really fast. I was frustrated.
So, I tried to understand why, and the reason was simple!
Agents don't properly know where to put code (they need clear rules). In general, they don't know how to architect code. You have to guide them.
Thus, I revised my classics: MVC, Clean Code, Hexagon, etc.
And, took the best from these and made a modern spec that fits modern apps' needs (scalability, iterating fast, etc).
Check it here: https://www.zapstudio.dev/specifications/uaa
Comments URL: https://news.ycombinator.com/item?id=47151334
Points: 1
# Comments: 0
AI-Driven Smart Contract Vulnerability Research
Article URL: https://github.com/CoinFabrik/scout-audit-ai/blob/master/docs/State-Of-The-Art%20Survey%20Document.md
Comments URL: https://news.ycombinator.com/item?id=47151330
Points: 1
# Comments: 0
Show HN: Me.txt – A personal identity file for AI agents
Article URL: https://www.metxt.org/
Comments URL: https://news.ycombinator.com/item?id=47151328
Points: 1
# Comments: 1
Gh-PR-reviews Fetch GitHub PR reviews as compact Markdown for LLM context
Article URL: https://github.com/Nittarab/gh-pr-reviews
Comments URL: https://news.ycombinator.com/item?id=47150891
Points: 1
# Comments: 1
From PHP Spaghetti to Elixir: A Real ERP Migration That Saved 35% on Ops
Article URL: https://www.destinationcrm.com/Articles/ReadArticle.aspx?ArticleID=173399
Comments URL: https://news.ycombinator.com/item?id=47150885
Points: 1
# Comments: 0
