21d ago
AI Agent Skill Traceability in UX Design
Title-only entry about UX design considerations for AI agent skill traceability.
21d ago
BareMetalRT – TensorRT-LLM running natively on Windows (no WSL)
BareMetalRT enables TensorRT-LLM and tool-calling agents natively on Windows GPUs with local-only data processing and auditing.
21d ago
Show HN: ContextNest versioned, governed context for AI agents (open-source CLI)
ContextNest open-source CLI providing versioned, governed context memory layer for AI agents to maintain persistent cross-session knowledge.
21d ago
Show HN: Tokenstead, find AI models for your hardware
Tokenstead tool for finding open-source AI models compatible with user hardware, with speed estimates and cloud pricing comparisons.
21d ago
Introducing Muse Spark 1.1
Meta Muse Spark 1.1 multimodal reasoning model optimized for agentic tasks with improvements in tool use, coding, and computer vision.
21d ago
A Startup That Builds AI Agents Used One to Raise $100M
News headline reporting startup using AI agent to raise $100M funding round.
21d ago
Ollama raises $65M, grows to nearly 9M users
Ollama raises $65M Series B, now 9M users; open-source tool enabling developers to run open-weight AI models locally on personal computers.
21d ago
Show HN: EvenKeel – a free financial planning chatbot
EvenKeel LLM-powered chatbot for financial planning using Gemini 3 Flash with skills library and Monte-Carlo simulations.
21d ago
Why AI Agents need a runtime, not just a framework
Technical analysis of why AI agent systems need runtime infrastructure for governance, versioning, auditing, and deployment beyond framework capabilities.
21d ago
Show HN: A zero-dependency auditor for decision records in AI-touched systems
Open-source zero-dependency auditor for AI-touched systems detecting governance gaps: missing owners, irreversible actions without gates, unwitnessed changes.
21d ago
A Prolog library for interfacing with LLMs
pllm is a SWI-Prolog library that integrates LLMs via HTTP endpoints, supporting OpenAI-compatible chat/completions APIs with configurable authentication.
21d ago
Engineering Sacrifice: How Open Source Survives the Age of Free Code
Zig and Godot projects ban AI-generated contributions, citing quality and review time concerns.
21d ago
Lab: The Full-Stack Platform for Training Your Own Models
Lab platform launches for agentic post-training research, unifying environments, training, and evaluations.
21d ago
Show HN: Magpie-search, Federated, local-first search for an AI
Magpie-search: federated local-first search engine for AI agents to maintain context across crashes.
21d ago
Reverse-engineering Nvidia's CUDA-checkpoint for faster cold starts
Reverse-engineering NVIDIA's cuda-checkpoint feature for faster GPU process startup and restoration.
21d ago
Show HN: FableCut – A browser video editor AI agents can drive (zero deps)
FableCut browser-based non-linear video editor with JSON timeline that AI agents can programmatically control via MCP/REST, zero npm dependencies.
21d ago
Building the AI Retrieval Infrastructure Behind 20B+ Vectors at HubSpot
HubSpot's semantic search infrastructure for 20B+ vectors powering RAG and AI agents across 38+ teams.
21d ago
Show HN: Virena, a minimal vision-language-action robot model you can read
Vision-language-action robot model trained from scratch for pick-and-place tasks. Author discovered observation quality matters more than architecture.
21d ago
LLMs for technical editing: The good, the bad, and the ugly
Opinion piece on limitations of using LLMs for technical editing roles. Author argues against replacing human editors.
21d ago
Show HN: Slopera, a browser that hallucinates every page with an LLM
Slopera: browser that renders web pages as LLM hallucinations instead of fetching real content. Uses Anthropic/OpenRouter APIs.
21d ago
Show HN: Noema – AI personas debate US listed stocks
AI personas debate US stock opportunities. LLM-based investment analysis tool providing multi-perspective reports.
21d ago
Show HN: OpenTab – a lazygit-style TUI for your AI token spend
OpenTab: TUI and web UI for tracking AI token spending across coding tools. Reads local SQLite databases from OpenCode and Claude Code.
21d ago
Show HN: Yogen – 500 AI agents argue about your idea before you bet on it
Yogen: simulates 500 AI agents debating ideas across 6 rounds with opinion cascading through social graph. Self-hosted option available.
21d ago
Show HN: Wizard, Self-extending autonomous AI agent in one Rust binary
Wizard: single-binary autonomous AI agent in Rust. Self-extending, supports multiple providers (local/API), browser automation via MCP.
21d ago
Show HN: Arcaide – Explore code with multi-level call graphs
Arcaide: visual code exploration tool with multi-level call graphs for navigating codebases at architectural level.
21d ago
Which Predictor? How GEPA Optimizes a Multi-Agent DSPy Program
Deep dive on GEPA optimization framework for multi-agent DSPy programs, exploring how to optimize components with multiple predictors.
21d ago
Show HN: Agent scans 130 careers pages nightly, scores jobs vs. your resume
AI agent that scans 130+ company careers pages nightly, scores jobs against user's resume using LLM, and drafts applications. Open source on PyPI.
21d ago
Build a Free AI Agent That Takes Actions (No Code)
No-code guide to building an AI agent that calls real REST APIs during conversations to answer questions with live data.
21d ago
Show HN: A control-theory approach to detecting LLM agent instability
Lyapunov-stability monitor for multi-turn LLM agents that detects token spirals and failure patterns without extra LLM calls, addressing 41-87% failure rates in production multi-agent systems.
21d ago
Show HN: Agent Sessions – local history and a live quota meter for Codex/Claude
Desktop app for browsing, searching and resuming local AI agent sessions across Claude, Codex, Cursor and other tools. Open-source with CLI support.
21d ago
AI and Job Postings: From Destruction to Creation?
Analysis of software development job postings showing rebound correlated with introduction of agentic AI tools like Claude Code.
21d ago
Show HN: SoulOS–State&persona management for agents without giving up your LLM
State and persona management system for AI agents without LLM dependency. Docker-based local deployment with offline documentation.
21d ago
Viability of Local Models for Coding
Technical exploration of viability of running local LLMs for coding tasks, comparing recent local model performance against cloud alternatives.
21d ago
Show HN: QX Labs – build agents, flows and grids that work across your tools
Platform for building composable AI agents with three primitives: agents for chat across Slack/WhatsApp/email, flows for workflows, grids for parallel processing.
21d ago
Show HN: WhisperShortcut – voice layer for AI on macOS (BYOK, offline Whisper)
Voice-first AI tool for macOS using offline Whisper transcription, clipboard integration, and support for Gemini/GPT/Grok APIs.
21d ago
Bypassing the hyperscaler AI PaaS tax with a Rust orchestration plane
Architecture discussion on using Rust orchestration to bypass hyperscaler AI PaaS costs for sovereign AI infrastructure.
21d ago
Ask HN: I hate coding agents. Is this skill issue?
Discussion of developer experience challenges when using AI coding agents, including context switching and loss of project ownership.
21d ago
LLM-as-a-Verifier: A General-Purpose Verification Framework
General-purpose verification framework using LLMs as verifiers. arXiv submission page without full content.
21d ago
V2 of Holographic Horizon Shield: LLM Defenses with Phi-3 Integration
LLM safety framework using geometric principles with Phi-3 integration for security boundary detection.
21d ago
Show HN: Ember – Lightweight headless browser for AI agents (17MB idle)
Open source lightweight headless browser for AI agents using 17MB idle memory, no Docker or API keys required, runnable on VPS/laptop/Raspberry Pi.
21d ago
Astryx by Facebook: An design system that's customizable and agent ready
Meta's open-source design system built for customization and agent collaboration. React and StyleX based.
21d ago
Keppsake – automatic save points for your AI projects
Tool for automatic checkpointing of AI projects on macOS, allowing rollback when AI generation goes wrong.
21d ago
Show HN: LangDrift – test AI agents across languages
Testing framework for evaluating AI agent behavior consistency across languages, checking tool selection and arguments in multiple locales.
21d ago
The true cost of saying "Hi" to an AI agent
Analysis of hidden costs in AI agent interactions, calculating human waiting time overhead alongside token costs.
21d ago
Show HN: Agentic FC – a football management SIM played by AI agents over MCP
Football management simulation where AI agents control teams through MCP, with terminal-based match commentary and ASCII visualization.
21d ago
Show HN: Internationalizationstack – Fix AI Localization Bugs in Claude Code
CLI tool suite and Claude Code skills for fixing AI localization bugs, providing format engineering, QA testing, and native-quality translations for 46 languages.
21d ago
Chasing the Hallucinations: KPMG's AI-Powered Attempt at "Redefining Excellence"
Investigation using hallucination detection tool to uncover fabricated citations in AI-generated government reports, academic papers, and consulting firm research.
21d ago
97% of websites expose zero tools an AI agent can use
Analysis of nearly 10,000 websites showing 97% expose no tools for AI agents to use, highlighting gap between agent-ready interfaces and current web design.
21d ago
ChatGPT is now a partner for your most ambitious work
ChatGPT Work agent that takes cross-app actions and manages multi-hour projects autonomously.
21d ago
Relm – local LLMs as base-R objects, with interpretability
R package with Rust core embedding local LLMs, exposing generation, embeddings, activation tracing, steering and ablation as base-R functions for mechanistic interpretability research.