Data Science and Analytics · Mountain View, CA

Staff Machine Learning Engineer, Infrastructure

Waymo · Mountain View, CA · Posted Dec 8, 2025

This posting comes from Waymo. Confirm the details on their site before applying.

Overview

Join our ML Infrastructure engineering team advancing state-of-the-art ultra-realistic multi-agent simulations using foundation models. In this role, you will work at the intersection of ML infrastructure, foundation models, and simulation engineering, with a specific focus on writing high-performance business and simulation logic in JAX/TensorFlow running directly on TPUs to power realistic environments for Reinforcement Learning (RL).

Responsibilities

  • Design, build, and optimize realistic simulation environments and business logic running on TPUs using JAX and TensorFlow. Implement and optimize large-scale model and data parallelism strategies for training and running foundation models on TPU hardware.
  • Collaborate closely with modeling teams to integrate foundation models into simulation pipelines.
  • Drive technical architectures and system designs from data engineering through simulation execution to meet business and performance objectives.
  • Profile systems, identify performance bottlenecks across ML accelerators, and optimize end-to-end execution speed.
  • Translate product and business goals into concrete technical requirements and system deliverables.

Requirements

  • 6+ years of professional software engineering experience, with at least 4 years focused on machine learning infrastructure (scaling, training, optimizing, and deploying large-scale ML systems).
  • Direct ML programming experience on TPU and GPU hardware using frameworks such as JAX, PyTorch, or TensorFlow.
  • Proven hands-on experience scaling large models using model parallelism, data parallelism, or distributed training techniques.
  • Strong understanding of state-of-the-art ML models (e.g., autoregressive transformers) and hands-on proficiency with ML accelerator profiling tools to diagnose bottlenecks.
  • Demonstrated ability to independently lead ambiguous technical initiatives end-to-end and build robust libraries, pipelines, and developer tooling.
  • Strong verbal and written communication skills to collaborate effectively across distributed, cross-functional teams.

Preferred qualifications

  • Practical experience in Reinforcement Learning (RL), Sim2Real transfer, or Robotics.
  • Experience with distributed ML frameworks and accelerators like GPU/TPU.
  • Domain familiarity with Autonomous Driving systems, multi-agent simulations, or realistic world modeling.

Benefits

  • Bonus

Great fits start here.

Whether you're looking for your next role or your next team member, give people and possibilities a place to meet.