Restaurant owner built a $120 camera + vision model system that cut table bussing time in half
bennash · x · 2026-11-14
A restaurant owner shares his DIY AI monitoring setup: three $40 cameras, OpenCV + YOLOv8 for person/tableware detection, a vision model via API classifying events and priorities as JSON, and a Telegram bot routing alerts by role. The system prompt includes the shift schedule, menu, closing checklist and a table-numbered floor plan.
It catches plates sitting too long under heat lamps, uncleaned tables, un-greeted walk-ins, open fridge doors, checkout queues, and back-door openings after hours. After three months: table bussing time dropped from 6 to under 3 minutes, greeting speed tripled, escalations to the owner fell 40%, and the monthly cost is less than one assistant-manager shift. Employees initially disliked it, then found alerts reached them before the boss noticed.
More from coding & agent
- Qwen 3.5 4B scores 10/10 on hidden coding tests fully local on a 16GB MacBook Air — Ok-Brush8377 · 2026-11-14
- HQ rebuilds local bot runtime: self-host agents, chat from desktop, Slack or Telegram — jacob_posel · 2026-11-14
- HQ adds MCP signup: create an account inside ChatGPT, Claude or Grok — jacob_posel · 2026-11-14
- Coolify TLS cert rotation works, but CA API buckles under eMudhra fallout rush — tristanbob · 2026-11-14
- Delta editor adds Claude Agent via ACP, works with existing Claude subscriptions — dean_rie · 2026-11-14
- Claude Code 2.1.324 is about to be released — ClaudeCodeLog · 2026-11-14