AI/ML Platform MLOps Engineer

Role overview Build and operate a cloud-based platform that supports the full machine-learning lifecycle, from experimentation and training through deployment, monitoring and retraining. The work combines platform engineering, MLOps automation, Python development and integration with high-performance computing (HPC), with a focus on secure, reliable production services.

Responsibilities - Design and run machine-learning infrastructure on Google Cloud, using Vertex AI for pipelines, model management and serving. - Provision reusable cloud infrastructure with Terraform and connect it to automated delivery workflows. - Build CI/CD pipelines with GitHub Actions, including secure identity and secrets handling. - Develop and manage Slurm-based HPC workloads, including GPU scheduling and workflows spanning cloud and HPC environments. - Create production Python services and REST APIs, and support model training, evaluation, deployment and retraining. - Monitor model performance, data drift and system health; improve reliability, traceability and platform usability with partner teams.

Requirements - Production experience in MLOps or machine-learning platform engineering. - Hands-on experience with Google Cloud and Vertex AI, plus Terraform for infrastructure as code. - Strong Python skills and experience with TensorFlow, PyTorch or scikit-learn. - Familiarity with MLflow and the KFP SDK, and with the end-to-end machine-learning lifecycle. - Experience building production APIs with FastAPI or Flask, and understanding OAuth 2.0 and OpenAPI specifications. - Experience with GitHub Actions, Workload Identity Federation and Slurm/HPC environments, including GPU scheduling. - Knowledge of distributed training, model deployment and secure cloud engineering practices.

Nice to have - Experience designing enterprise-scale ML platforms or hybrid cloud/HPC architectures. - Knowledge of data governance, compliance or model risk management. - Experience improving GPU workload performance and cloud costs. - Background in scientific, analytics or data-intensive environments.

The role calls for independent ownership, clear documentation and planning, careful engineering, and effective collaboration with data science, product, platform and DevOps colleagues.