ML Compute Efficiency Automation Engineer, Infrastructure & Planning

What you'll do

• Build systems of record for resource requests, allocations, and utilization to govern compute as code.
• Identify and fix inefficiencies in ML inference and training workloads across GPUs, TPUs, and Apple Silicon.
• Solve core optimization problems regarding scheduling, capacity allocation, and serving costs.
• Automate manual workflows, triage, and reporting to eliminate operational toil and manual escalations.
• Develop telemetry, schemas, and anomaly detection to surface efficiency and cost opportunities.
• Re-architect processes that buckle under scale to ensure they grow with usage instead of headcount.
• Create reusable tooling to provide long-term infrastructure support for the broader team.

What we're looking for

• BS in Computer Science, Computer Engineering, or equivalent practical experience.
• 6 or more years of experience building production software, automation, tooling, or data and infrastructure systems.
• Strong programming skills in Python or similar languages for automation, pipelines, and tooling.
• Proficiency with SQL and experience creating dashboards or data products in tools like Tableau, Looker, or Grafana.
• Experience designing data models or telemetry schemas for infrastructure, capacity, or utilization data.
• Experience running complex systems in large-scale compute, cloud, or infrastructure environments.
• Production experience shipping automated or autonomous workflows (preferred).
• Understanding of ML training and inference infrastructure, GPU/TPU utilization, and foundation model serving (preferred).