Research Engineer, ML Infrastructure
Core
Build and own distributed training infrastructure, experiment orchestration, and data pipelines to support large-scale AI research and agent rollouts.
Role type
Senior IC ML Infrastructure Engineer
Builds
Distributed training systems, agent rollout infrastructure, experiment orchestration tools, and high-throughput data pipelines
Domain
AI/ML Infrastructure, Distributed Systems, GPU Computing
Deliverable
production ML models
Required skills
Distributed systems, GPU performance profiling, Python, C++, PyTorch, Parallelism strategies (data/tensor/pipeline/sequence), Systems debugging, Data pipeline engineering
Preferred skills
Experience with large model training, Research tooling development, Understanding of ML research workflows
Technologies
PyTorch, Python, C++, GPU clusters, VM sandboxes
Responsibilities
Build distributed training infrastructure for large-scale GPU jobs; Optimize training throughput and memory utilization; Design experiment orchestration and tracking systems; Build high-throughput data pipelines for training and evaluation; Debug complex distributed system failures; Implement parallelism strategies for large models; Anticipate and build infrastructure needs ahead of research demands
Seniority
Senior, hands-on IC