About
The foundation is there. We’re working with existing vision APIs, but we need someone who can make the system actually work at the level we need. That means integrating the API, tuning and calibrating it against our specific taxonomy, building the processing pipeline around it, handling edge cases, and ensuring classification accuracy stays high as we scale.
What you’d be doing:
Integrating and configuring vision APIs for multi‑label classification across 70+ categories
Fine‑tuning and calibrating model outputs to match our specific content taxonomy
Building the end‑to‑end pipeline ingestion, processing, tagging, and validation
Handling video processing (frame sampling, scene detection, aggregating predictions)
Monitoring accuracy, identifying failure patterns, and continuously improving results
Building tooling to make it easy for the team to review, correct, and refine classifications
#J-18808-Ljbffr
Languages
- English
Notice for Users
This job comes from a TieTalent partner platform. Click "Apply Now" to submit your application directly on their site.