All projects
yyzmeet.sh
yyzmeet_case_study.sh
~/projects cat yyzmeet_case_study.txt
// Problem
Recording Google Meet audio without installing software or paying for enterprise recording tools requires automation that joins meetings and stores artifacts reliably.
// What I built
I built YYZ Meet as a standalone public Google Meet audio recorder. Users paste a Meet URL or open /{meet-code}, admit guest "YYZ Recorder," then download audio and captions via a 6-digit key when the meeting ends.
// Architecture
FastAPI orchestrator manages job lifecycle with SQLite index. Docker worker containers run Playwright for browser automation and ffmpeg for audio extraction. Private AWS S3 stores artifacts with presigned download URLs. 6-digit key system for secure artifact retrieval.
// Stack
Python FastAPI Docker Playwright ffmpeg AWS S3 SQLite
// Outcomes
- ▸ Standalone public Google Meet audio recorder at yyzgoogle.com
- ▸ Public GitHub repo with Docker-based Meet recording pipeline
- ▸ Automated audio + caption extraction via Playwright and ffmpeg workers