Member of Technical Staff - ML Ops
Core
Build and scale the data generation pipeline and job execution platform for a foundation model that learns physics from simulations to enable physical reasoning in engineering design.
Role type
Senior ML Ops Engineer (Simulation Data & Infrastructure)
Builds
A petabyte-scale data corpus of validated physical simulations and the job scheduling/execution platform that generates it.
Domain
AI/ML infrastructure for physics simulation and semiconductor design
Deliverable
production ML models
Required skills
Python, distributed execution frameworks (Temporal, Ray), workflow orchestration, GPU cluster management, data pipeline scaling, database management (relational/NoSQL), observability
Preferred skills
PyTorch distributed training, scientific data (CFD, FEA), Infrastructure as Code, parameter sampling strategies
Technologies
Temporal, Ray, Kubernetes, Ray Train, torchrun, Prometheus, OpenTelemetry
Responsibilities
Scale data generation pipeline from thousands to millions of simulations; own dataset lineage and reproducibility; build data quality checks into the pipeline; run and extend the GPU job platform; close the loop between data generation and model training; integrate with infrastructure team; optimize compute costs; maintain codebase health with tests and documentation
Seniority
Senior, hands-on IC