Senior Software Engineer, Post-Training & RL Frameworks

The Waymo ML Frameworks & Efficiency team partners with Research and Production teams across Waymo to develop models in Perception and Planning that are core to our autonomous driving software. We help our partners by offering the best frameworks for the entire model development lifecycle, including pre-training and post-training. They are geared towards both scaling models with efficiency and solving problems unique to ML for autonomous driving.

We are looking for engineers with ML system expertise to help us train and improve pre-trained models to be deployed into Waymo Driver, and potential future products. You'll work with researchers and modeling engineers across the company, and solve the challenges of large-scale reinforcement learning (RL), building systems that can scale across compute, data, and environments to improve model intelligence and understanding of human drivers.

You Will:

• Report into the Head of ML Frameworks & Efficiency
• Develop the core training system for adapting RL techniques to unprecedented scales and heterogeneous environments (i.e. CPU/GPU/TPU).
• Collaborate with teams to integrate the latest rollout strategies, policies, and RL algorithms (i.e. REINFORCE, DPO, PPO) into the system.
• Improve the end-to-end RL training pipeline for efficient and scalable learners/actors, and low-latency distributed reply buffers for persisting data produced by the rollouts.
• Build evaluations, analyze experimental results and iterate quickly to improve model performance and training workflows.
• Stay current with the latest research in RL, Vision-Language-Action (VLA) models, and World models to inform and inspire new programs.

You Have:

• B.S. in Computer Science, Math, or 8+ years equivalent real-world experience.
• Proficient in distributed systems design with an understanding of ML efficiency.
• Experience with ML frameworks, including TensorFlow, JAX, XLA.
• Solid programming skills in Python and C++.
• Practical familiarity with profiling tools to uncover performance bottlenecks.

We Prefer:

• MS in Computer Science, Math

Familiarity with post-training frameworks like TS/REX, Tunix, TorchRL, TRL.

#LI-Hybrid