Chef

Overview

Chef is an AI app builder that “knows the backend”: it ships with a built-in database, zero-config auth, file uploads, real-time UIs, and background workflows so developers can rapidly build full-stack AI applications. Chef uses Convex as its backend for real-time and persistent data, making it well suited for generative applications that need backend capabilities.

Key Features

• Backend integration: Tight integration with Convex provides realtime data and background workflows.
• Agent loop and tooling: The chef-agent folder contains agent loops, system prompts, and tool definitions to build intelligent assistant flows.
• Hosted and local development: Run the hosted app at chef.convex.dev or run locally for development and testing.

Use Cases

• Generative apps that need backend storage and realtime sync (code generation platforms, collaborative editors, stateful chatbots).
• Rapid prototyping to production: use templates and the chefshot CLI to bootstrap projects.
• Educational and demo scenarios showing how agent capabilities integrate with DB, auth, file uploads, and realtime UI.

Technical Characteristics

• TypeScript monorepo with client and server code for modern frontend integration.
• Template-driven project bootstrapping (template/) for quick project initialization.
• Lightweight CLI and test harnesses (chefshot, test-kitchen) to support local development and agent-loop testing.