Machine Learning DevOps Engineer – Cloud and Compute Cluster

Stack

Responsibilities

• ▹Optimize infrastructure for ML training and inference (GPUs, distributed compute)
• ▹Automate and maintain ML/LLM pipelines: data ingestion, training, validation, deployment
• ▹Manage model versioning, reproducibility and traceability
• ▹Work with terabyte-scale datasets
• ▹Implement ML-centric CI/CD practices
• ▹Monitor model performance and data drift in production
• ▹Collaborate with machine learning engineers, software engineers and platform teams

Requirements

• ▹Very good familiarity with Linux, shell scripts and cluster configuration
• ▹Proficiency in workload management, containerization and orchestration (Slurm, Docker, Kubernetes)
• ▹Solid grasp of CI/CD tools and workflows (GitHub Actions, Jenkins, GitLab CI)
• ▹Cloud infrastructure knowledge (AWS, GCP, Azure), especially ML services (SageMaker HyperPod, Vertex AI)
• ▹Familiarity with monitoring/logging tools (Grafana, CloudWatch, Prometheus, Loki)
• ▹Experience with infrastructure as code (Terraform, CloudFormation, cluster-toolkit)
• ▹Experience with ML pipeline orchestration tools (MLflow, Kubeflow, Airflow, Metaflow)
• ▹Python programming skills with ML libraries (TensorFlow, PyTorch)
• ▹Experience with cluster, systems and network administration
• ▹Minimum BSc in Computer Science or Information Technology

Nice to have

• ▹Accepted contribution to the Linux kernel
• ▹Won an important bug bounty
• ▹Support of an academic grid/cluster compute infrastructure

Soft skills

About the company

Pathway builds a post-transformer AI architecture (BDH), trusted by organizations such as NATO, La Poste and Formula 1 teams, headquartered in Palo Alto and led by renowned AI researchers.