ML Platform Engineer, Training Platform
Tencent
- Location
- Shenzhen
- Level
- Senior
- Posted
- 26 Sep 2026
SeniorOn-siteActive
Original posting
Skills
- ML platform
- Distributed training
- Kubernetes
- Python
- GPU
Summary
Tencent's Shenzhen training-platform team is hiring an ML platform engineer. Modeling teams depend on this platform for GPU jobs, experiment tracking, and failure recovery. You will not own a model metric. You will improve scheduling and retry behavior for multi-node jobs, and debug a stuck run with the algorithm engineer who submitted it. The role is on site in Shenzhen. Requirements: ML platform or distributed-systems experience, Python or Go, and Kubernetes. Familiarity with how a training job fails is more important than a publication.
Full description
Tencent's Shenzhen training-platform team is hiring an ML platform engineer. Modeling teams depend on this platform for GPU jobs, experiment tracking, and failure recovery. You will not own a model metric.
You will improve scheduling and retry behavior for multi-node jobs, and debug a stuck run with the algorithm engineer who submitted it. The role is on site in Shenzhen.
Requirements: ML platform or distributed-systems experience, Python or Go, and Kubernetes. Familiarity with how a training job fails is more important than a publication.