Lead Machine Learning Engineer

Essential Responsibilities

• Design, train and evaluate state of the art models for May’s autonomous driving, simulation and ML Platform stack.
• Leverage emerging techniques in the End-to-End driving, Vision Language Action (VLA), World or Foundation model domains to solve commercial-scale problems.
• Lead small teams of cross functional Engineers beyond the state of the art.
• Define data balance, training experiment and evaluation practices to train efficiently at petabyte scale.

Skills and Abilities

Success in this role typically requires the following competencies:

• Direct experience architecting & training VLA, MMLM, or Generative World Models for commercial-scale applications
• Experience composing, processing and characterizing large (>100TB) multi-modal datasets
• Experience analyzing and addressing long-tail failure cases in large models
• Experience leading teams of 2-3 Engineers and communicating technical details to interdisciplinary leadership.

Qualifications and Experience

Candidates most successful in this role typically hold the following qualifications or comparable knowledge or experience:

Required

• Extensive practical experience in one of the following domains:
  • Vision Language Action Models
  • Generative World Models
  • Foundation Models in Robotics
  • Data Centric AI
• A minimum of 4 years of industry experience working on commercial robotics systems.
• A minimum of 1 year mentoring ML Engineers in a commercial or lab environment.
• Master’s degree in Robotics, Computer Science, or Computer Engineering, or a field that requires a strong mathematical and/or engineering foundation.
• Practical experience handling the “Long Tail” problem in Machine Learning.
• Strong programming skills in Python/PyTorch in a Linux environment.
• Functional understanding of LiDAR, Camera and Radar processing techniques.

Desirable

• PhD and/or published research in the described specialty domains.
• Familiar with common post-training techniques.
• Experience deploying models to resource constrained and edge hardware
• Functional understanding of C/C++/CUDA memory and threading models