Platform Engineer
Core
Build and maintain the ML infrastructure platform enabling AI researchers and materials scientists to train, evaluate, and deploy models at scale.
Role type
Senior Platform Engineer (MLOps/Infrastructure)
Builds
Cloud-native ML infrastructure on Kubernetes, CI/CD pipelines, model registries, and distributed training systems.
Domain
AI/ML Infrastructure, Cloud Computing, Materials Science
Deliverable
production ML models
Required skills
Kubernetes, Python, Go, Terraform, Helm, GitOps, Distributed ML Training (JAX/PyTorch), GPU Cluster Management, Infrastructure as Code, Multi-cloud (AWS/GCP)
Preferred skills
Flyte, Kapitan, Pants, HPC environments (Slurm/PBS), ML Serving (Triton/TorchServe/KServe), Multi-tenant platforms
Technologies
Kubernetes, Terraform, Helm, Kapitan, GitOps, AWS, GCP, JAX, PyTorch, Flyte, Pants, Triton, TorchServe, KServe, Slurm, PBS
Responsibilities
Design and implement cloud-native platforms on Kubernetes for model training and deployment; Implement and maintain full MLOps stack including CI/CD and experiment tracking; Build infrastructure for distributed GPU training and resource optimization; Ensure platform reliability through monitoring and automated recovery; Manage complex multi-environment configurations; Optimize cloud costs and computational throughput; Create developer tools for researchers; Collaborate with ML researchers and scientists to build infrastructure.
Seniority
Senior, hands-on IC
