Senior AI Infrastructure Engineer - Training Platform
$216k - $270k • San Francisco, CA; Seattle, WA; New York, NY
Posted 2mo ago
About the job
As a Software Engineer on the Machine Learning Infrastructure team, you will build the "Operating System" for our large-scale GPU clusters. You will architect a high-performance training platform that handles the immense complexity of multi-thousand GPU workloads, ensuring every cycle is used efficiently. Your work directly determines the velocity at which our researchers can train and iterate on the world’s most advanced models. The ideal candidate is a systems expert who thrives on solving the orchestration, networking, and reliability challenges that emerge at massive scale. You will partner closely with researchers to build a seamless, resilient environment that transforms raw compute into breakthrough AI.
Responsibilities
- Architect and scale a multi-tenant orchestration layer for GPU clusters, ensuring high utilization and seamless job recovery.
- Design and implement scheduling primitives to optimize the lifecycle of training jobs.
- Develop deep observability and automated health-checking into the training stack to proactively identify and isolate hardware failures.
- Evaluate and integrate emerging technologies in the CNCF and AI ecosystem, making data-driven build vs. buy decisions.
- Work closely with Finance and Procurement teams to drive capacity planning.
- Participate in on-call rotations to ensure service availability.
- Own projects end-to-end, from requirements and scoping to design and implementation.
Requirements
- 5+ years of experience in backend or infrastructure engineering, with at least 2 years focused on orchestrating ML workloads at scale (100+ GPU nodes).
- Strong programming skills in Python, Go, Rust, or C++.
- Experience with complex compute management systems covering queueing, quotas, preemption, and gang scheduling.
- Experience with distributed training infrastructure (e.g., EFA, Infiniband, topology-aware scheduling).
- Experience with distributed storage systems (e.g., Lustre, S3) related to training throughput.
- Expert-level knowledge of Kubernetes internals (Custom Resources, Operators, Admission Controllers) and device plugins.
- Familiarity with cloud infrastructure (AWS, GCP) and infrastructure as code (e.g., Terraform).
- Proven ability to solve complex problems and work independently in fast-moving environments.
Benefits
- Base salary
- Equity
- Comprehensive health, dental, and vision coverage
- Retirement benefits
- Learning and development stipend
- Generous PTO
- Commuter stipend