Senior Software Engineer, Applied ML and Evaluation

The Perception Scene Semantics team is responsible for developing ML models for scene understanding around construction, emergency scenes, mapless driving and world changes. The team works on the end to end ML pipeline from data mining, labeling, training onboard / offboard models, and deployment. We need someone with strong ML experience to drive the architecture upgrades and product feature development.

In this hybrid role, you will report to a TLM within Perception Semantics.

You will:

• Design and implement new evaluation systems and strategies to measure the safety and performance of large-scale Waymo Perception machine learning models and support onboard engineers in both regression prevention and quality hillclimbing.
• Make evaluation-driven recommendations for modeling and data changes and actively influence deployment of onboard models.
• Create evaluation strategies that leverage advanced ML strategies and models, like VLMs, active learning

You have:

• 2+ years of experience in industrial AI applications involving the creation, maintenance, and evaluation of ML products
• Ability to drive the development of new products
• Experience in JAX (Preferred), TensorFlow (Preferred), or PyTorch
• Functional knowledge of C++ programming
• Passion for developing novel ways of measuring the performance of mission-critical ML systems
• Experience evaluating the quality of machine learning prediction at scale

We prefer:

• Background in both computer vision and general machine learning.
• Experience using and developing distributed systems
• Good knowledge of LLM/VLM

#LI-hybrid