Product Radar
A unified feed of emerging software, developer tools, and early-stage market launches.
NUA, an AI agent that generates product correctness tests by interpreting company context, including laws and regulations.
An AI agent designed to overcome the limitations of existing AI-generated tests by focusing on 'product correctness' and 'intent,' particularly for non-technical bugs in regulated industries.
Uindow
The programmable browser with trusted interactions
VAEN, an open-source CLI for packaging and importing AI coding-agent harnesses.
A solution to the problem of non-portable AI coding-agent workflows, enabling easy sharing and extraction of complete agentic harnesses (instructions, skills, mcp servers) via a single .agent file.
A browser-based gaming platform for 'core gamers,' featuring WASM ports of classic games and indie Unity titles, with a future focus on WebGPU titles, aiming to be a new discoverability layer for smaller developers.
A browser-based alternative to platforms like Steam for core gamers, leveraging WASM and WebGPU to deliver console/PC quality titles directly in the browser. It is positioned as a solution for developer discoverability outside crowded channels.
Life Navigator
Become The Best Version Of Yourself
An open-source agentic QA harness with memory.
An open-source tool for agentic Quality Assurance, emphasizing memory capabilities.
SLayer, an open-source semantic layer designed for AI agents to interact with databases, enabling dynamic model exploration, query execution, schema editing, and learning from interactions.
An 'agent-native' semantic layer that overcomes the limitations of traditional BI-centric semantic layers and raw SQL for agentic workflows, allowing agents to iterate, learn, and evolve the data layer.
An archived "Old School UFO Website."
A historical, content-rich UFO website praised for its UI compared to contemporary government releases.
LaoTzu Writer Studio – a modern writing application with AI elements.
A writing tool that helps "build the framework of a novel" and provides AI support (e.g., flagging inconsistent canon, analyzing writing stats) but explicitly "leaves the writing for you to do," distinguishing itself from AI story generators.
Pathetic – a high-performance, low-allocation Java pathfinding library
A Java pathfinding solution that achieves C++-like performance, enabling 10k concurrent paths in ~7ms and handling large path distances without heap bloat, by addressing fundamental Java performance issues like object-heaviness and GC-hostility.
Agent-evals – Claude skill to build your own evals
A practical starting point for evaluating AI agents, specifically for startups lacking data science expertise, by leveraging a Claude Skill to set up evaluation baselines directly in the codebase.
Kanban-CLI – a web UI for local Markdown todo lists
A CLI utility with a web UI for managing agentic tasks using local Markdown files, aimed at reducing external SaaS spend and consolidating context within a monorepo.
Ragnerock, an AI data analysis tool for automating data wrangling and processing.
A comprehensive platform for data scientists to automate LLM-driven data processing and analysis, offering customizable pipelines, a unified SQL query interface, Jupyter-compatible notebooks, and integration with various AI models, databases, and web crawling.
Nimbus, a desktop browser with an integrated AI agent and a redesigned user experience.
A desktop browser reimagined with an AI agent as the primary interface, eliminating traditional browser chrome elements like the URL bar in favor of a chat-based interaction model.
Should – Expressive Assertions for Go
A Go testing library providing expressive assertions, designed to immediately clarify test failures without requiring debug prints.
MemFactory: Unified Inference and Training Framework for Agent Memory
The first unified, highly modular training and inference framework specifically designed for memory-augmented agents, abstracting the memory lifecycle into plug-and-play components. Integrates Group Relative Policy Optimization (GRPO) for fine-tuning memory management policies.
no-mistakes, a local Git proxy that validates code via an AI coding agent pipeline before forwarding to the remote, automating PR creation and CI babysitting.
A method to "kill AI slop" by enforcing validation and automating development workflow steps.
Open Passkey, an MIT licensed open-source passkey repository, and Gateway, a free "backendless" hosted authentication server with an end-to-end encrypted key-value store.
An easy, secure authentication solution for rapid deployment of small frontend applications, eliminating backend configuration and improving developer velocity. Positioned as a superior alternative for secure API key storage without server costs.
XitDB – an immutable single-file database.
Fills a gap for an immutable (like Datomic) and embedded (like SQLite) database, offering a unique combination of features for data integrity and simplified deployment.
Lazyagent – A TUI for monitoring AI coding agents
A terminal-based UI that provides comprehensive observability for multiple AI coding agents (Claude Code, Codex, OpenCode), addressing the complexity of tracking agent activities, subagent spawns, tool calls, and code changes across different runtimes and projects.
SaaS Metrics
Hacker News
Product Hunt