Senior Deep Learning Software Infrastructure Engineer
Core
Build and scale training libraries and infrastructure enabling end-to-end autonomous driving models on thousands of GPUs.
Role type
Senior Deep Learning Software Infrastructure Engineer
Builds
Training libraries, orchestration frameworks, and fault-resilient systems for massive GPU clusters.
Domain
Autonomous Vehicles / Deep Learning Infrastructure
Deliverable
production ML models
Required skills
Distributed training (DDP/FSDP), multi-GPU orchestration, Python library development, datacenter networking (RoCE/IB), parallel filesystems (Lustre), performance profiling, Kubernetes/Slurm
Preferred skills
Fault resilience, high availability, elastic training, large-scale observability, leadership in ML systems engineering
Technologies
PyTorch, NCCL, Kubernetes, Slurm, Lustre, RoCE, IB
Responsibilities
Craft and harden deep learning infrastructure libraries for multi-thousand GPU clusters; improve efficiency in data loaders, distributed training, and scheduling; build robust pipelines for massive video datasets; collaborate with researchers to minimize training stalls; own core infrastructure components like orchestration and fault-resilient systems; partner with leadership to scale infrastructure with growing GPU capacity.
