Overview
AI Coffee Manager is an autonomous business-analyst agent deployed on Google Cloud Run. It reads a coffee shop's historical point-of-sale data from a Google Sheet, writes and runs its own Python analysis code inside a Cloud Run sandbox, and diagnoses demand patterns such as graduation-weekend spikes. The agent proposes staffing and inventory fixes, but it is designed not to write to the shop's live spreadsheet until a manager explicitly approves in chat.
The project is implemented as a single Cloud Run service with a FastAPI and WebSocket chat UI. It uses Google Agent Development Kit (ADK) components, Gemini via Vertex AI, and Google Sheets API v4. The agent's workflow includes ingesting POS-2025 data, correlating last year's per-hour demand spikes with this year's ceremony schedule, diagnosing bottlenecks, proposing a task list, and acting only after a "yes" reply.
The README also lists limitations: in-memory sessions reset if the Cloud Run instance restarts or scales to zero; user_id and session_id are fixed for a single active session; the manager currently pastes the ceremony schedule into chat; and general Q&A about the coffee menu is not tool-grounded.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Coffee shop managers use AI Coffee Manager to review historical POS data and recommend staffing and inventory preparation for graduation weekend.
- Managers use it to identify whether a predicted high-volume slot is limited by cashier shortage or barista/complex-drink shortage.
- Managers use it to add approved tasks to a TODO-2026 tab without manual spreadsheet editing.
- Operators use it to execute agent-written Python analysis code in a Cloud Run sandbox.
- Managers use it to receive staffing recommendations such as scheduling an additional Support Barista.
- Managers use it to receive inventory recommendations for Cold Brew, espresso bean, and alternative milk stock.
- Developers use it to deploy the agent from source on Cloud Run with buildpacks and a least-privilege service account.
Features
Analyzes coffee shop POS data
Predicts graduation-weekend demand spikes
Recommends staffing and inventory adjustments
Runs analysis in a Cloud Run sandbox
Requires approval before writing to Sheets
Offers a live WebSocket chat interface
Reviews
No reviews yet. Be the first to share your experience!