Hi Product Hunt, I'm Titouan, and I built Factorai. Coding agents are now good enough to write most of the code. Claude Code and Codex (also all other agents) take a task, read the codebase, make the change and run the tests. The bottleneck has moved from writing code to orchestrating the agents that write it: which one is working, which one is waiting on you, what each one changed, and whether you can trust it. Our tools did not move with them. The IDE is still built for one human, one cursor and one file at a time, with the agent squeezed into a terminal at the bottom of the screen. Factorai is built the other way round. The session is the unit of work: - Each session is the real claude or codex CLI, in a real terminal that keeps running when you look away. - A dot says whether it is working, waiting for you, or stopped. - The diff, the commit graph and search across every session sit next to the terminal, so reviewing the agent's work never means leaving the app. - Routines run a saved prompt on a schedule, in the background. Why fully open source and community-driven? The way we build software is being redefined right now, and the workbench for it should not belong to one vendor or sit behind a subscription. Every engineer deserves good tooling for working with agents, whatever their company or budget, and the people using it should decide where it goes. So Factorai is MIT licensed and local-only, with no account and no telemetry. There is no company and no paid tier behind it, and none planned. It runs on macOS, Linux and Windows through WSL 2. It is still early, but several senior engineers use it every day. Fixes land fast on the alpha channel, and stable gets them once alpha users have run them. What I would love from you: try it on a real project and tell me where it gets in your way. If you want to help shape it, issues and pull requests are open on GitHub, and I read every one.

Factorai

IDE is dead. Long live the ADE (Agentic Development Env)

IDE is dead. Long live the ADE (Agentic Development Env)