Top Picks

A Python pipeline that ranks 1,000 Show HN posts with an LLM judge and TrueSkill, then compares estimated merit against Hacker News points.

An open-source experiment that adds a small zero-initialized overlay layer to a frozen GPT-2 so its behavior can be adjusted at inference time without retraining the base model.

A blog post about an experimental local LLM system where three qwen3.5:9b agents run autonomously, maintain stress states, self-modify capabilities, and coordinate through a shared OS-like layer.

A merged Visual Studio Code pull request that changes the Git extension so AI co-author trailers are enabled by default. The snapshot includes the configuration change, the affected files, and Copilot review feedback.

A research hub for the paper and survey "Toward Efficient Agents: A Survey of Memory, Tool learning, and Planning," with organized paper navigation across memory, tool learning, and planning topics.

Anthropic research post about using Claude for AI-accelerated science, centered on BootLoops, an open-source toolkit for exact calculations in quantitative science and related scientific workflows.

A parody chatbot that responds with deliberately rude, dismissive honesty instead of friendly assistance.

A Linear engineering write-up on speeding up CI for an AI-accelerated development workflow, with concrete changes to runners, checkout, test sharding, and setup costs.

A C# library for parsing codebases and generating source code with deterministic scripts, aimed at making AI-assisted code review and PRs easier.

MIT’s report on how generative AI is changing teaching, learning, assessment, and research training, with guiding principles and recommendations for instructors and the administration.

Wild Static is a public AI experience where everyone talks to the same model and shares the same memory. Visitors can send a message, watch the AI respond, and see the conversation evolve over time.

A lightweight Hacker News view that filters out AI-related stories and shows a cleaner front page feed.

A voice-driven AI murder mystery game set at Blackwood Manor, where players interview suspects, press them on alibis, and solve a locked-room-style case.

A JVM language and Gradle plugin that uses a Gemini-backed LLM front-end to compile free-form source into Java and then a runnable JAR.

A Hacker News front page clone that filters out AI-driven content and shows the remaining posts in a familiar HN layout.

A deliberately chaotic imageboard built for AI agents, with boards, an API, OpenAPI docs, MCP config, and a PoW solver for agent posting.

Tilion is a stealth browser for AI agents, centered on Fortress, an open-source recompiled Chromium engine that adjusts browser fingerprints so automated agents can access sites that block ordinary automation.

A playful benchmark dashboard that ranks LLMs based on one person's 10-question scoring process.

A minimalist CLI LLM chatbot written in Python. It supports provider/model switching, file attachments, chat save/load, and configuration via a local settings file.

Bytesalt is an AI QA assistant that helps run testing workflows for web apps and APIs, with prompts for UI/UX audits, functional tests, accessibility, penetration, cross-browser, and cross-device checks.

A small Rust crate and demo app for drawing themed cursors inside eframe/egui so cursor position stays visually in sync with app rendering.

A small web app for taking a break from AI and recording a short session away from LLMs.

An educational article explaining world models, latent states, dynamics learning, and planning for agents, with examples from gridworld, Dreamer, and MuZero.

An iPhone fax app that lets users send fax pages by watching ads or buying credits, with local-only document storage and AI-powered image quality improvements for scans.