🛠 Tech Stack

💼 About This Role

You'll scale and optimize training systems and core model code, owning critical infrastructure for large-scale training. You'll work with researchers and model engineers to translate ideas into experiments and production runs, at the intersection of ML, software engineering, and scalable infrastructure.

🎯 What You'll Do

• Design and maintain systems for large-scale model training scheduling and job management
• Scale distributed JAX training across TPU and GPU clusters
• Profile and optimize memory usage, device utilization, and throughput
• Build abstractions for launching, monitoring, and debugging experiments

📋 Requirements

• Strong software engineering fundamentals with experience in JAX or PyTorch
• Hands-on large-scale training experience
• Familiarity with distributed training, multi-host setups, and evaluation pipelines
• Experience managing training workloads on cloud platforms like Kubernetes or SLURM

✨ Nice to Have

• Deep ML systems background (training compilers, runtime optimization)
• Experience with GPU/TPU performance tuning
• Background in robotics or multimodal models