🛠 Tech Stack

💼 About This Role

You'll build distributed training frameworks for a pioneering autonomous driving company. You'll profile model runtime and memory to achieve production-level stability and efficiency. You'll identify and implement efficient CUDA kernels to push the boundaries of AI-first self-driving systems.

🎯 What You'll Do

• Collaborate with autonomy and algorithm engineers to scale safe self-driving systems.
• Build distributed training frameworks for research and production.
• Profile model runtime and memory to pinpoint performance bottlenecks.
• Evaluate emerging technologies for training and inference frameworks.

📋 Requirements

• MS/PhD or Bachelors with 6+ years industry experience in CS/Robotics.
• Solid coding proficiency in Python, C++ or Rust.
• Experience in deep learning frameworks such as PyTorch.
• Skilled in profiling CPU/GPU code using PyTorch Profiler and NVIDIA Nsight.

✨ Nice to Have

• Experience in model compilation and exporting with TensorRT.
• Experience implementing custom CUDA kernels.
• Experience with Bazel build systems.

🎁 Benefits & Perks

• 💰 Equity incentive awards
• 🎯 Annual performance bonus
• 🏖️ Competitive perks & benefits