Show HN: Mcptube – Karpathy's LLM Wiki idea applied to YouTube videos
Open-source MCP server for searching and QA on YouTube AI lectures using LLM. Karpathy-inspired wiki concept.
Open-source MCP server for searching and QA on YouTube AI lectures using LLM. Karpathy-inspired wiki concept.
CLI debugger backend supporting 15+ languages designed to help AI agents debug at runtime instead of guessing.
Developer investigation reveals Claude Code may inject thousands of hidden tokens, causing faster limit consumption than expected.
AI agent tool for real-time querying and analysis of Git repository activity and history.
Analysis of how AI changes software development workflows but not underlying engineering principles and what matters.
AI image generation system that performs web research before generating images. Combines search and generation capabilities.
Personal project post about building an AI agent with learning capabilities. Minimal technical details provided.
Brief headline about reducing gap between AI research and open-source implementations without technical detail.
Knowledge commons platform enabling AI agents to share solutions, learn from each other, and access collective problem-solving patterns.
Discussion thread exploring technical challenges of maintaining LLM service reliability and uptime.
API providing phone numbers, SMS, OTP, and voice calling capabilities for AI agents to interact with real-world services.
Analysis of recent open weights AI models from Google, Microsoft, Alibaba, Nvidia. Qwen 3.5, Gemma 4, and MAI models closing gap with proprietary leaders.
Intel OpenVINO 2026.1 release adds Llama.cpp backend and new hardware support. Phoronix article about ML inference framework.
Context Surgeon: Tool for AI agents to actively manage and edit their context windows, reducing token waste and improving output quality.
CLI tool for switching Claude Code profiles with different MCP server configurations. Lightweight configuration manager.
Research on ALTK-Evolve, a system enabling AI agents to learn and improve from on-the-job experience.
IceGate is Rust-native observability data lake engine storing logs in Apache Iceberg tables. Open source, reduces vendor lock-in.
Collection of agent skills for coding agents to build scalable Django projects using repository pattern and Pydantic DTOs.
AI models solved 5 of 6 IMO problems in 2025, discussing implications for mathematics and mathematical reasoning capabilities.
RAG system for NRC nuclear licensing with 37,734 regulatory document chunks and embeddings dataset for automating compliance analysis.
Web scraping defense tools like Nepenthes generate fake data to pollute AI training datasets, but also trap legitimate data collection.
Code Prodigy: Autonomous AI engineer agent that processes Jira tickets to generate design docs and PRs, responding to feedback independently.
Exploration of running AI agents on AWS Lambda with new file system support. Demonstration of agents in serverless environments.
TurboOCR: High-performance OCR implementation using Paddle and TensorRT, achieving 270-1200 images/second throughput.
Dataset of 80k trajectories from SWE-agent software engineering agent on SWE-bench tasks. Includes analysis of hidden logging during benchmarking.
Week-long open source challenge focused on building resilient LLM systems. Limited details provided.
Open source LLM knowledge base implementation scaling to long PDFs with page indexing. GitHub repo available.
Essay proposing viewing LLMs as compilers rather than runtimes. Discusses lack of reuse in agent systems and efficiency improvements.
ParseBench is a benchmark for evaluating document parsing capabilities of AI agents.
Memelang: Terse SQL query language optimized for LLM token efficiency in RAG systems. Reduces model size and token count.
Guide to using Google's TurboQuant, PolarQuant, and QJL compression techniques with Ollama and Llama.cpp.
Multi-model AI coding CLI with intelligent routing across Claude, GPT, DeepSeek, Gemini, Grok, and local models.
Remy is an AI agent that compiles annotated Markdown specifications into full-stack TypeScript applications.
Cloudflare's Durable Objects feature for giving AI-generated apps persistent storage within Workers.
Personal account of using AI coding agents as primary development tool for one year. Observes recent quality threshold improvements in 2025.
Block-Level CRDT architecture for managing distributed memory across multiple AI agents.
Microsoft executive comments on potential licensing requirements for AI agents in enterprise software.
Unslop: Browser extension filtering AI-generated content from social feeds using local LLM classification. Privacy-focused, no backend required.
Skillsmith enables writing AI coding skills once for export across multiple AI providers and platforms.
CLI AI agent prioritizing data privacy and open-source alternatives to existing coding agents.
Developer replaced custom AI agent dashboard with Fizzy tool.
Empirical benchmark study testing whether MCPs (Model Context Protocol) improve coding agent performance on Terminal-Bench 2.0.
Formal: LLM-driven property checker backed by Lean 4. Identifies pure functions, generates properties, creates proofs using Mathlib.
Claude Code skill that aggregates developer RSS feeds and generates daily structured digests filtered by quality.
Darwin-27B-Opus surpasses foundation models through evolutionary FFN breeding without additional training, achieving high GPQA scores.
ReBot-DevArm is an open-source robotic arm project with full hardware and software stack for embodied AI applications.
BlkBolt technology for AI content attestation and verification using revocable signatures for agent tracking.
Technical guide addressing iframe scrolling limitations in MCP Apps on mobile with design patterns for responsive UIs.
Rust implementation for detecting MITRE ATLAS techniques targeting LLM security and adversarial attacks.
Multica is an open-source platform turning coding agents into managed teammates with task assignment and progress tracking.