Build and operate machine learning systems on GCP. You will package and serve models through containers and KServe or Vertex endpoints, and keep them reliable in production with monitoring and drift detection.
The work spans feature and model registries, CI/CD, cost-efficient autoscaling, GPU scheduling, canary and shadow deployments, and A/B testing for model releases.
Job Description
We are looking for experienced AI & ML Engineers with strong MLOps responsibilities and expertise in Python, AI/ML and cloud technologies.
Required Skills and Experience:
-
Python
-
PyTorch
-
Understanding of transformer architecture
-
Agent architectures such as ReAct
-
GCP knowledge
-
Experience of Docker and cloud databases
-
RAG
Additional Technical Requirements:
-
Skilled in packaging and serving models using containers and KServe/Vertex endpoints.
-
Experience working with feature and model registries, CI/CD, model monitoring, and drift detection.
-
Experience with cost-efficient autoscaling, GPU scheduling, canary and shadow deployments, and A/B testing.
- Package and serve models using containers and KServe/Vertex endpoints.
- Maintain feature and model registries alongside CI/CD pipelines.
- Monitor models in production and detect drift.
- Implement autoscaling, GPU scheduling, and canary or shadow deployments.
- Support A/B testing of model releases.
- Python and PyTorch skills.
- Understanding of transformer architecture and agent architectures such as ReAct.
- GCP knowledge with Docker and cloud database experience.
- Experience with RAG.
- Experience with model registries, CI/CD, model monitoring, and drift detection.