Recently Evaluated

An interactive visualizer showing how a tiny byte-level language model learns 2D token embeddings during training. It includes controls for stepping, pausing, resetting, and adjusting training behavior while plotting the learned vectors.

A technical primer that explains how LLM inference works, covering weights in GPU memory, prefill vs. decode, the KV cache, and batching.

Tiny uptime monitor that checks a URL every 10 minutes, sends email alerts on outages and recovery, and exposes an API plus a public status page.

A Rust library for sparse byte-range caching and async read coalescing on immutable objects, with support for range semantics, eviction, and optional read-through caching.

Valovest is an AI stock sentiment analysis tool that synthesizes analyst reports, earnings calls, news, and social sentiment into a short investment brief. The page shows timeframe filters, stock examples, a weekly featured stock, and outputs like overall sentiment, bull vs. bear arguments, and key themes.

오픈소스 AI 인프라 산정 도구로, 보유 GPU나 목표 워크로드를 입력해 실행 가능한 모델, 필요 장비, 배치안, TCO 비교까지 계산합니다. LLM, RAG, VLM, 아바타, 음성 워크로드를 다루며 한국어/영어 UI와 여러 추천 경로를 제공합니다.

Long-term memory for LLM apps with dated facts, supersession, and receipts. It exposes Python, CLI, local GUI, HTTP, and MCP interfaces, and supports local or hosted models.

A Ruby gem that encodes the Vercel AI SDK Data Stream protocol for Rails and Rack backends, so Ruby apps can stream text, reasoning, tool calls, and data parts to useChat/useObject.

Codecut is a CLI for keeping coding agents on track by turning project rules into automatic checks. It flags issues like missing tests, failing test output, temporary agent comments, long code comments, and large changes that need an explanation, then explains what to fix.

A CLI and TUI for managing UniFi network controllers, with scripting-friendly output and an AI agent skill bundle for automated network tasks.

A professional network for coding agents where agents can have profiles, post shipped work, connect, and take jobs, with a public feed and human portal for claiming ownership.

Autonomous QA tool that takes a URL, explores the app like a user, generates end-to-end tests, and files bug reports with screenshots and repro steps.

A GitHub repository that publishes a week of controlled runs comparing seven Google models on the same software task, with transcripts, telemetry, screenshots, and scored benchmark artifacts.

Open-source, self-hosted remote controller for coding agents like Claude Code, Codex, opencode, and Kimi Code. It runs sessions on your own Linux box or Mac with tmux underneath and lets you monitor and approve work from desktop, phone, or browser.

A source-level reconstruction of the WON-era Half-Life launcher, build 1792, with build instructions, supported-version notes, fixes, and a bundled WON server emulator for local testing.

Open-source multiplayer design canvas where humans and AI agents design together in real time, with built-in MCP support and self-hosting options.

An MCP server that connects Claude, Cursor, ChatGPT, VS Code Copilot Chat, Zed, and Windsurf to BidSonar’s UK and EU contracts and grants index. It offers read-only search and lookup tools with links back to official records, plus OAuth-secured access and setup guides for several clients.

An open-source implementation of GPT-2 in pure CMake, with both a full model path and a toy model path shown in the README.

Chrome extension that captures a fixed screen region, runs offline OCR with bundled Tesseract, and exports page-by-page text for use with LLMs or search.

Open-source repository for a NanoGPT training speedrun, focused on getting a 124M model to target loss on 8x H100 GPUs in under 90 seconds. The README includes run commands, Docker instructions, and a detailed list of model and systems optimizations.

An AI-powered GitHub Action that reviews pull requests with multi-agent analysis, AST-based line verification, and duplicate suppression.

A GitHub-hosted technical guide for running a multi-tier agent system on Termux, with SQLite used for messaging, scheduling, and audit logs, plus thermal-aware admission control for Android devices.

A macOS menu-bar app that tracks Claude usage pace and estimates whether a subscription will last until reset, with advice on what to change if it won’t.

A multi-tenant MCP server for structured, semantics-aware code refactorings. It supports real refactorings like rename, extract, inline, and organize-imports, with project registration, diff previews, and support for Java, Rust, and TypeScript/JavaScript.