Vinci
Member of Technical Staff - ML Ops
$185K–$245K
Palo Alto HQ · Hybrid · Staff · Full-time · Posted Oct 30, 2026
- Experience
- 5+ years; includes at least 2 years owning a GPU pipeline at meaningful scale.
About the job
Build and operate the simulation-data and ML job platform that produces validated, traceable datasets and connects them to model training and deployment.
What you'll do
- Scale simulation data generation and build dataset lineage, reproducibility, and quality checks.
- Extend the GPU job platform and own training launches, checkpoint management, and the model registry.
- Improve workload reliability and cost visibility while coordinating compute and data needs with infrastructure.
What they ask for
- 5+ years building production data or ML pipelines, including at least 2 years owning a GPU pipeline at meaningful scale.
- Experience running large batch or simulation workloads with failures, retries, and preemption.
- Strong Python and software engineering fundamentals, including typed code and testing.
- Experience with relational and NoSQL databases at scale, workflow orchestration or distributed execution, and cloud object storage.
Nice to have
- PyTorch distributed training, checkpointing, and experiment tracking.
- MPI, CUDA or HIP, numerical solvers, scientific or simulation data, infrastructure as code, or observability stacks.
Pay
$185K – $245K (base salary; currency not specified)
Tools
Python, Temporal, Ray, Ray Train, torchrun, Airflow, Argo, Kubernetes Jobs
AI focus
Build the data and ML lifecycle infrastructure for a foundation model that learns physics from simulations, including training runs and paths to reuse or deploy trained checkpoints.
Model-written summary · the original posting is authoritative