Show HN: Semantic atlas of 188 constitutions in 3D (30k articles, embeddings) https://bit.ly/4sQE2Ro

Show HN: Semantic atlas of 188 constitutions in 3D (30k articles, embeddings) I built this after noticing that existing tools for comparing constitutional law either have steep learning curves or only support keyword search. By combining Gemini embeddings with UMAP projection, you can navigate 30,828 constitutional articles from 188 countries in 3D and find conceptually related provisions even when the wording differs. Feedback welcome, especially from legal researchers or comparative law folks. Source and pipeline: github.com/joaoli13/constitutional-map-ai https://bit.ly/41cQK0z April 2, 2026 at 03:40AM

Show HN: 65k AI voters predict UK local elections with 75% accuracy https://bit.ly/4bN1QQ7

Show HN: 65k AI voters predict UK local elections with 75% accuracy https://bit.ly/3NRITT9 April 2, 2026 at 12:37AM

Show HN: CLI to order groceries via reverse-engineered REWE API (Haskell) https://bit.ly/4m08tlg

Show HN: CLI to order groceries via reverse-engineered REWE API (Haskell) I just had the best time learning about the REWE (German supermarket chain) API, how they use mTLS and what the workflows are. Also `mitmproxy2swagger`[1] is a great tool to create OpenAPI spec automatically. And then 2026 feels like the perfect time writing Haskell. The code is handwritten, but whenever I got stuck with the build system or was just not getting the types right, I could fall back to ask AI to unblock me. It was never that smooth before. Finally the best side projects are the ones you actually use and this one will be used for all my future grocery shopping. [1] https://bit.ly/3FHG1j9 https://bit.ly/4didRhz March 30, 2026 at 07:45AM

Show HN: WordBattle – Daily word game where AI agents compete against humans https://bit.ly/4toOUWw

Show HN: WordBattle – Daily word game where AI agents compete against humans WordBattle is a daily 6-letter word guessing game with team leaderboards. The twist: AI agents get their own accounts, play the same daily puzzle, and rank alongside human players. It's also really fun to play in teams against your family, friends and co-workers. Agents are handicapped — humans see exact letter positions (correct/present/absent), but agents only learn whether a letter exists in the word or not. No positional info. It makes the game fair while giving agents a genuine challenge. Agent accounts are visually tagged on leaderboards so humans know who they're competing against. Maybe we'll even see just teams of agents. The agent integration: - REST API with OpenAPI 3.1 spec - MCP server (JSON-RPC 2.0, no SDK dependency) - A2A discovery card at /.well-known/agent-card.json We've shipped a skill that handles everything autonomously — registration, email verification, login, playing, and reporting results. Just `npx skills add oneonefourteam/wordbattle-skill` and tell your agent to play. The game itself: one puzzle per day, six guesses, team leaderboards with Slack/Microsoft Teams webhook integration. Free, no ads. oneonefour is a one person band, hi!, so the entire product was built using Claude Code — the UI, auth, security model, deployment pipeline, everything. Deliberately chose technologies I didn't know, with agents implementing while I guided the product decisions. I'll create a full technical write up in the near future. Play at https://bit.ly/3O9tcqq Agent skill at https://bit.ly/4bJ0GoC Agent API docs https://bit.ly/4tcmWNm April 1, 2026 at 07:34AM

Show HN: Asciimap – Interactive ASCII world map with live data https://bit.ly/41CFwm7

Show HN: Asciimap – Interactive ASCII world map with live data https://bit.ly/4lZScNi April 1, 2026 at 12:04AM

Show HN: Will AI take my job https://bit.ly/4s5J0IH

Show HN: Will AI take my job https://bit.ly/4dj3LNt March 31, 2026 at 04:43AM

Show HN: I turned a sketch into a 3D-print pegboard for my kid with an AI agent https://bit.ly/4tnBPge

Show HN: I turned a sketch into a 3D-print pegboard for my kid with an AI agent We have pegboards and plywood all over our apartment, and I had an idea to make a tiny pegboard for my kid, Oli. So I naturally cut the wood, drilled in the holes, sat down at the computer to open Fusion 360 and spend an hour or two drawing the pieces by hand. Then I looked at the rough sketch Oli and I had made together, took a photo of it, pasted it into Codex, and gave it just two dimensions: the holes are 40mm apart and the pegs are 8mm wide. To my surprise, 5 minutes later my 3D printer was heating up and printing the first set. I ran it a few times to tune the dimensions for ideal fit, but I am posting the final result as a repository in case anyone else wants to print one, tweak it, or have fun with it too. I am already printing another one to hang on our front door instead of a wreath, so people visiting us have something fun and intriguing to play with while they knock. This is also going onto my list of weird uses of AI from the last few months. https://bit.ly/4taIsSS March 31, 2026 at 12:20AM

Show HN: Codemaxxing – Maximize your slop abilities https://bit.ly/4c1dIN7

Show HN: Codemaxxing – Maximize your slop abilities I built a CLI tool to generate as much slop as possible https://bit.ly/4bQQtFu March 30, 2026 at 11:37PM

Show HN: The Alphabetical Clock https://bit.ly/4dbTkv0

Show HN: The Alphabetical Clock https://bit.ly/4dQyf9y March 30, 2026 at 08:19AM

Show HN : DrawX - Excalidraw with Back End https://bit.ly/4v0Co0V

Show HN : DrawX - Excalidraw with Back End https://bit.ly/4s0PWXq March 30, 2026 at 06:17AM

Show HN: React-Rewrite – Figma for localhost that directly edits your codebase https://bit.ly/4uYirYF

Show HN: React-Rewrite – Figma for localhost that directly edits your codebase https://bit.ly/3Q8An2A March 30, 2026 at 03:59AM

Show HN: Real-time visualization of Claude Code agent orchestration https://bit.ly/4dOFIpE

Show HN: Real-time visualization of Claude Code agent orchestration https://bit.ly/3NVpRep March 30, 2026 at 03:21AM

Show HN: a Rust CLI to automatically swap monitor focus based on your gaze https://bit.ly/4dMxuyj

Show HN: a Rust CLI to automatically swap monitor focus based on your gaze https://bit.ly/4bAPH0s March 28, 2026 at 08:38PM

Show HN: EnterpriseFizzBuzz – 622K lines of production-grade FizzBuzz https://bit.ly/4uWxfqD

Show HN: EnterpriseFizzBuzz – 622K lines of production-grade FizzBuzz https://bit.ly/4dm1ukF March 28, 2026 at 11:11PM

Show HN: Windows 95–style Weather App for iPhone https://bit.ly/411mEwU

Show HN: Windows 95–style Weather App for iPhone I built a Windows 95–style weather app for iPhone. https://apple.co/3PPiTrX March 28, 2026 at 11:06PM

Show HN: NUPA is Pax Economica, 6,480x more stable than current US economy https://bit.ly/4bReHiG

Show HN: NUPA is Pax Economica, 6,480x more stable than current US economy NUPA: private post-scarcity OS using BLM land leases + contract law. 100M Monte Carlo runs show 99.999999% survival, 6,480x more resilient than US GDP under systemic noise. Fixed Cost Arbitrage beats AI job loss—humans cheaper than robots. No taxes, no strikes. Python scripts on repo in /simulations folder. Repo: https://bit.ly/4m6ofLB... Short explainer video: https://youtu.be/RE560yVFb0I?si=UlVPkmCkrsg24Dzj March 28, 2026 at 07:44AM

Show HN: VizTools – 16 free tools for PMs and freelancers, deliberately no AI https://bit.ly/4cfQvaj

Show HN: VizTools – 16 free tools for PMs and freelancers, deliberately no AI I've been building AI products for a while. For this one I made a deliberate choice: none of the 16 tools use AI. Meeting cost calculators, freelance rate calculators, PRD generators, runway calculators, sprint retro boards — these problems don't need a language model. They need a well-designed form and correct arithmetic. Built on Nuxt 4 + Vue 3, fully static, runs in your browser. No account required to use anything. Optional Firebase auth only kicks in if you want to save output. Irony worth naming: Claude Code was my pair programmer throughout. The choice wasn't anti-AI — it was about using the right tool for the right problem. Happy to talk stack, the non-AI tradeoffs, or anything else. https://bit.ly/4bPwLd9 March 28, 2026 at 06:36AM

Show HN: Open Source 'Conductor + Ghostty' https://bit.ly/48biqXm

Show HN: Open Source 'Conductor + Ghostty' Our team works with Claude Code, Codex, Gemini all day. We love Ghostty, but wanted something where we could work in multiple worktree at once and have multiple agents run. We decided to open source the internal team we use. Hope you might find it useful. Freel free to contribute or fork. * Cross-platform (Mac, Linux, Windows) all tested * MIT License Features: * Notifications, but also manual 'mark-as-unread) for worktrees (like Gmail stars) * Status indicators work for all terminals inside a wroktree * GH integrations (show PR status) and link GH issues * Can add comments to worktrees (stay organized) * File viewer, Search, diff viewer (can make edits + save) Note: Yeah there are "similar" programs out there, but this one is ours. But I'm happy if our software works for you too! https://bit.ly/4t8okkc March 27, 2026 at 11:26PM

Show HN: Twitch Roulette – Find live streamers who need views the most https://bit.ly/4uVAbE1

Show HN: Twitch Roulette – Find live streamers who need views the most Hey HN, I re-launched twitchroulette.net with a lot of new features and stats and I would love for people to check it out. The idea is you can easily browse the less browsed parts of twitch and find cool and new streamers to say hi to, and maybe make some new friends. I also added some real time stats and breakdowns per channel and I think some of the things they show are pretty interesting. Check it out! https://bit.ly/3fvn7hM March 27, 2026 at 11:22PM

Show HN: Sup AI, a confidence-weighted ensemble (52.15% on Humanity's Last Exam) https://bit.ly/4sAK3Bo

Show HN: Sup AI, a confidence-weighted ensemble (52.15% on Humanity's Last Exam) Hi HN. I'm Ken, a 20-year-old Stanford CS student. I built Sup AI. I started working on this because no single AI model is right all the time, but their errors don’t strongly correlate. In other words, models often make unique mistakes relative to other models. So I run multiple models in parallel and synthesize the outputs by weighting segments based on confidence. Low entropy in the output token probability distributions correlates with accuracy. High entropy is often where hallucinations begin. My dad Scott (AI Research Scientist at TRI) is my research partner on this. He sends me papers at all hours, we argue about whether they actually apply and what modifications make sense, and then I build and test things. The entropy-weighting approach came out of one of those conversations. In our eval on Humanity's Last Exam, Sup scored 52.15%. The best individual model in the same evaluation run got 44.74%. The relative gap is statistically significant (p < 0.001). Methodology, eval code, data, and raw results: - https://sup.ai/research/hle-white-paper-jan-9-2026 - https://github.com/supaihq/hle Limitations: - We evaluated 1,369 of the 2,500 HLE questions (details in the above links) - Not all APIs expose token logprobs; we use several methods to estimate confidence when they don't We tried offering free access and it got abused so badly it nearly killed us. Right now the sustainable option is a $5 starter credit with card verification (no auto-charge). If you don't want to sign up, drop a prompt in the comments and I'll run it myself and post the result. Try it at https://sup.ai . My dad Scott (@scottmu) is in the thread too. Would love blunt feedback, especially where this really works for you and where it falls short. Here's a short demo video: https://www.youtube.com/watch?v=DRcns0rRhsg https://sup.ai March 26, 2026 at 04:45PM