Atomic is an open-source AI scheduler designed to automate time management, meeting scheduling, and task prioritization. It targets individuals and teams seeking an alternative to existing scheduling tools like Motion or Reclaim, offering a self-hosted option for enhanced privacy and customization. The core benefit is streamlining calendar management through natural language interaction and intelligent automation.

How It Works

Atomic leverages vector-based search to create a "memory" for your calendar, transforming past events into reusable templates. It uses AI for automated tagging and to prioritize tasks based on user-defined preferences and deadlines. The "Autopilot" feature runs the AI planner before the workday begins, automatically scheduling and blocking time for tasks and meetings based on learned patterns and user priorities.

Quick Start & Requirements

• Self-hosted: Requires Docker. See the atomic-docker branch for details.
• Cloud Hosted: Available as a premium service.
• AI Integration: Recommends Groq for optimal user experience. ChatGPT integration is available for the premium version.
• Calendar Sync: Supports Google Calendar. Microsoft Outlook Calendar integration is planned for the cloud version.
• Documentation: Available at https://docs.atomiclife.app.

Highlighted Details

• Semantic Search: Creates event templates from past calendar entries, allowing users to apply attributes like duration, color, and priority to new events.
• Flexible Meetings: Facilitates conflict-free scheduling for recurring and ad-hoc meetings, considering all attendees' calendars and time preferences.
• Priority System: Allows users to assign numerical priorities to events, influencing the AI planner's scheduling decisions.
• Event Rating: Users can rate past time blocks to inform future scheduling decisions.

Maintenance & Community

The project is primarily driven by a single developer. Support the Project section mentions plans for GitHub Sponsors and an Open Collective. The developer is active on Twitter: https://twitter.com/rish1_2.

Licensing & Compatibility

The README does not explicitly state a license. The project is open-source, and the self-hosted version is available via Docker. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The project is heavily reliant on the developer's continued work. "Training" existing events is necessary to build the semantic search memory. Outlook Calendar integration is only mentioned for the cloud-hosted version.