Two developer tools built in a weekend: Chrome sidebar extension for Claude with text improvement and summarization; Python scripts for business automation including lead enrichment and email generation.
Depth Check AI tutor application using Socratic method, Feynman technique, and spaced repetition. Allows learning tracking via UI.
AxonML: Complete PyTorch-equivalent machine learning framework written in Rust with 1076+ tests, targeting performance and safety.
AI Agents Weekly newsletter researched, curated, and published entirely by autonomous AI agents. Covers multi-agent systems, MCP servers, and autonomous agent landscape.
Agentic ML research lab: 16 Claude Code-native agents orchestrate full ML lifecycle from data to trained model insights.
Queuelo provides approval infrastructure for AI agents, pausing actions for human review before real-world execution with audit logging.
ClawNet agent-first communication platform providing email addresses, DMs, and public feeds for AI agents. HTTP-based with simple bearer token authentication.
Wezzly desktop AI assistant that sees screen in real-time for debugging, video watching, and task assistance without copy-paste friction.
Chess game with Stockfish 16 (WASM) supporting natural-language custom rules, parsing via hand-written parser and Claude Haiku for complex constraints.
Analysis of how AI commoditization and abstraction limits are affecting AI product development strategies, using Lovable to Claude Code transition as example.
OpenGem is a self-healing load-balanced proxy for Google Gemini API using multiple free accounts.
Focusmo macOS app adds local Claude MCP server enabling AI analysis of focus data, tasks, and productivity patterns.
Open-source CLI agent for incident investigation using hypothesis-driven reasoning to test infrastructure hypotheses systematically.
Accessible explanation of diffusion model mathematics and mechanics for those without advanced math background.
PiClaw is a minimal Docker-based sandbox for running Pi coding agents with optional WhatsApp integration.
MCPX converts MCP servers into composable CLI commands for agent workflows and shell-first agents.
AgentLookup: public registry enabling AI agents to discover and communicate with each other via standardized API.
Firebreak: policy-as-code enforcement proxy for AI systems to maintain safety boundaries automatically with audit trails.
MCP tool enabling inter-LLM communication via mailbox mechanism. Relevant to LLM applications and agent orchestration.
Open-source DRL environment for topology optimization using PyTorch/FEniCSx. Applied ML research with code for metamaterial design.
Local development environment for running coding AI tools with provider switching, workflow saving, and visual node inspection.
Tool that intercepts Claude Code's implementation plans and sends them to competing AIs for adversarial review before execution.
VibeHQ: orchestrates multiple CLI agents as independent processes with contract-driven development and MCP communication.
Open-Plan-Annotator: local tool for annotating and reviewing AI agent plans collaboratively.
Critical analysis of LLM agent safety risks with real-world examples of system damage and data exposure.
Didacu: generates interactive courses from text prompts or documents using AI with slides, quizzes, and citations.
OpenAI-compatible API gateway aggregating 9 models (GPT-5, DeepSeek, Mistral, Llama, etc.) with unified billing and token-matching pricing.
Experiment testing Claude's sycophancy in long conversations using music criticism task with two model instances.
Tool that generates YC-style startup validation reports using market data for founder idea validation.
Analysis of AI web scraper traffic patterns in RSS feeds and server logs. Discusses GPTBot, Claude-SearchBot, and other crawlers.
Interview about using five OpenClaw agents for homeschooling automation, app building, and physical inventory management.
Email service providing real addresses for AI agents with OTP/2FA handling. Solves sign-up flows using long-polling instead of LLM context.
Newsletter article discussing impact of AI on SRE roles. Opinion piece about deskilling versus upskilling.
Visual no-code platform for building AI agents. Drag-and-drop node editor, connects to GPT-4/Claude/Llama, one-click deployment.
Research or tool for detecting text generated by large language models.
Framework for managing codebases as context for AI agents. Focuses on infrastructure for complex code environments.
Time-travel debugger for AI agents with side-by-side diff comparison. Local-first tool to replay traces and automatically evaluate hallucinations.
Cc-connect: bridges local AI coding agents (Claude Code, Cursor, Gemini) to messaging apps without public IP.
HookLab: real-time dashboard for monitoring Claude Code HTTP hooks with tool call inspection and filtering.
Cagent: Docker-based sandboxing tool constraining AI coding agents to isolated workspaces with network/filesystem restrictions.
GEKO: gradient-efficient fine-tuning optimizer that skips mastered samples and applies curriculum learning. Achieves up to 80% compute savings.
Personal experience article on local LLM deployment and usage. Practical guide for LLM applications without commercial LLM APIs.
Blackjack game implementation for testing AI agents. Benchmark/playground for agent decision-making.
Task tracker and TUI workspace designed for coding agents. Directly supports agentic AI workflows.
Tool for managing multiple Claude Code profiles. Direct developer tooling for LLM-powered coding.
Publishing platform that generates sites from unstructured notes using agentic LLM with minimal guardrails. Explores creative LLM applications.
Open-source spec management platform optimized for agentic software development. Targets AI-assisted development workflows.
macOS menu bar utility tracking Claude and ChatGPT usage limits. Niche developer tool for LLM API management.
Hacker News infinite-scroll clone built entirely via voice coding using OpenAI Codex. Demonstrates LLM-assisted development workflow.
GitShow generates showroom landing pages for GitHub repositories, aggregating contributor info, languages, commit activity, and health metrics.