I built Friday because I wanted an AI assistant that could do more than answer questions in a browser. Friday runs directly on your desktop and can work with files, search the web, run commands, draft content, generate images, video and audio, handle voice conversations, and automate recurring tasks. You can also extend it with reusable skills and MCP servers. A few principles shaped the product: • Local-first: settings, credentials, memory, and conversation history stay on your device. • Bring your own AI: choose your own providers and models for chat, voice, images, video, and audio. • Permissioned actions: Friday asks before writing files, sending messages, accessing private information, or performing other sensitive actions. • Cross-platform: available for macOS, Windows, and Linux. • Extensible: connect local or remote MCP servers and build reusable workflows. Friday is currently in beta, and I’m especially interested in feedback about the onboarding experience, the actions you would trust a desktop AI assistant to perform, and the integrations you would find most useful. Thanks for checking it out—I’ll be here throughout the launch to answer questions and hear your feedback.

Friday

Your Desktop AI Personal Assistant

Your Desktop AI Personal Assistant