Research Scientist / Engineer – Training Infrastructure

Highlights

• Build systems training large-scale multimodal models for physical-world intelligence.
• Create monitoring and debugging tools that keep massive training runs reliable.
• Enable researchers to focus on innovation atop reliable, scalable infrastructure.

Ideal candidate

A systems-minded engineer who thrives on hard scaling problems and loves making massive training runs stable and efficient so researchers can innovate faster.

Requirements

• Extensive distributed PyTorch training and parallelisms in foundation-model training.
• Deep understanding of GPU clusters, networking, and storage systems.
• Familiarity with communication libraries (NCCL, MPI) and distributed-system optimization.