Software Engineer Ii, Behavior Planning ML Platform

Aurora Innovation Aurora Innovation · Robotics · PITHQ · Software Engineering

Software Engineer II, Behavior Planning ML Platform at Aurora Innovation. This role focuses on building and optimizing the ML platform for self-driving technology, including offline pipelines for data extraction, model training, and evaluation, as well as onboard infrastructure for model deployment and inference. Requires BS degree, C++/Python skills, and ML framework experience.

What you'd actually do

  1. Develop large scale pipelines for data extraction, model training and model evaluation
  2. Build and optimize onboard ML infrastructure used to deploy models and run inference onboard the vehicle
  3. Collaborate closely with motion planning, systems engineering, and other autonomy groups to define and develop critical ML workflow requirements.

Skills

Required

  • BS or higher degree in Computer Science/Engineering or related fields
  • 6 months of experience
  • Strong programming skills in C++ or Python
  • Experience with machine learning frameworks (PyTorch or TensorFlow)
  • Solid foundation in computer science fundamentals - especially operating system concepts including concurrency, memory management and process scheduling.

Nice to have

  • Experience with GPU programming and Nvidia profiling tools - CUDA, NSight Systems, Nsight Compute
  • Experience working in cloud compute (AWS, Azure, GCP) ideally with batch processing or other large-scale workflow management platforms.
  • Experience profiling and optimizing software on Linux

What the JD emphasized

  • onboard ML infrastructure
  • run inference
  • large scale distributed pipelines
  • train and evaluate machine learning models

Other signals

  • building ML infrastructure
  • onboard ML infrastructure
  • run inference
  • large scale distributed pipelines
  • train and evaluate machine learning models