Launching today

Bernstein

Launching today

Spawn parallel AI agents. Ship tested code.

3 followers

Spawn parallel AI agents. Ship tested code.

3 followers

Visit website

Most AI coding tools give you one agent, one task, one answer. Bernstein gives you many. Set a goal. Bernstein breaks it into tasks, spawns Claude Code, Codex CLI, and Gemini CLI in parallel, verifies every result with your tests, and commits what passes. You come back to working code. Zero LLM tokens on coordination — the orchestrator is pure Python. No framework to learn. No vendor lock-in. Mix models in one run.

Free

Launch tags:Developer Tools•Artificial Intelligence•GitHub

Launch Team / Built With

ProAI — All-in-one AI Advisor, Forecaster, Creative, & Fundraiser

All-in-one AI Advisor, Forecaster, Creative, & Fundraiser

Promoted

Maker

📌

I built Bernstein because I was tired of babysitting AI agents one at a time. Claude Code is great. Codex CLI is great. Gemini CLI is great. But you can only talk to one at a time — watch it work, wait, then start the next task. Meanwhile you're the bottleneck. The insight: agents don't need to talk to each other to work in parallel. Give each one a clear task, a fresh git worktree, and a test suite. Let a deterministic orchestrator verify what passes. That's it. Bernstein is pure Python — zero LLM tokens on coordination. It auto-detects whatever agents you have installed (Claude, Codex, Gemini, Aider, Amp...) and mixes them in one run. Cheap free-tier models for boilerplate, heavy models for architecture. Clean git history when done. Happy to answer anything — what would you use it for?

Report

19h ago

Forum Threads

p/bernstein

•

18h ago

I ran 12 AI agents for 47 hours straight. Here's what broke (and what shipped).

Hey Product Hunt! I'm Alex, maker of Bernstein.

Last month I ran 12 AI coding agents on a single laptop for 47 hours - Claude Code, Codex CLI, Gemini CLI working in parallel. 737 tickets closed. 826 commits. 15.7 tickets per hour.

Everything that could go wrong did: merge conflicts, context drift, runaway token costs, agents overwriting each other's work, worktrees missing node_modules, port collisions, agents stuck in infinite loops burning through API credits.

Bernstein is every fix I built during that sprint, packaged into an open-source tool:

p/bernstein

•

18h ago

Hot take: single-agent coding tools are the new FTP

Using one AI agent at a time in 2026 feels like deploying via FTP in 2015.

It works, but you're leaving 80% of the throughput on the table.

I've been running 5-12 agents in parallel for the past month. The productivity difference isn't incremental - it's a step change. One agent on auth, one on tests, one on docs, one on API endpoints. All at the same time.

View all