Hey Product Hunt! 👋 I built Vibe Sensei because using AI for software development still felt surprisingly sequential. I was constantly switching between Claude, Codex, Gemini, terminals, repositories, and browser tabs—briefing one agent, waiting, reviewing the result, and then repeating the process with another. The agents were fast, but I had become the bottleneck. Vibe Sensei changes that workflow. It is a native desktop cockpit where an orchestrator can command multiple AI coding agents in parallel. Each agent runs in its own visible terminal, works inside an isolated Git worktree, and returns structured handoffs when its task is complete. You can: • Run Claude, Codex, Gemini, Cursor, and other AI CLIs side by side • Create reusable missions and agent squads • Watch every command and intervene at any time • Track tokens and costs by agent and mission • Dictate commands using local Whisper transcription • Preserve project context with persistent workspace memory • Use your own provider accounts without token resale Vibe Sensei runs locally on your machine. Your source code, tokens, and voice recordings remain under your control. This is the tool I wanted for my own development workflow, and I’m excited to finally share it with the Product Hunt community. I’d love to hear how you currently coordinate multiple coding agents—and which integrations or workflows you would like us to build next. 🥋

Vibe Sensei

AI coding agents in parallel, Your cockpit for AI coding

AI coding agents in parallel, Your cockpit for AI coding