Audio/Video Processing Prototype

We are building a prototype that processes audio and video data and generates structured outputs (e.g. speaking time, activity levels, simple lesson analysis).

The focus is on practical implementation, not research.

You will build :
- processing of audio/video input (files or streams)
- integration of existing tools:
+ speech-to-text (e.g. Whisper)
+ speaker diarization
- calculation of simple metrics:
+ speaking time per speaker
+ silence / overlap
- generation of structured output (JSON / API)
- simple backend (FastAPI)

Tech stack (indicative)
- Python
- FastAPI
- FFmpeg
- Whisper (or similar)
- PostgreSQL (optional)
- Docker (nice to have)

Profile
- 2–5 years of experience with Python
- experience with backend/API development
- experience with audio/video processing is a plus
- able to work independently from clear specifications
- pragmatic and solution-oriented

Practical
- freelance / part-time or full-time
- remote
- start asap
- duration: 4–8 weeks (initial phase)

This is not an AI research role. You will use existing tools and focus on building a working system.

To apply please include:
- relevant projects
- experience with Python/APIs
- short explanation of how you would approach this technically
- availability