About the job
Waymo is at the forefront of autonomous driving technology, committed to becoming the world's most trusted driver. Originating from the Google Self-Driving Car Project in 2009, Waymo has dedicated itself to developing the Waymo Driver—The World’s Most Experienced Driver™—to enhance mobility access while significantly reducing traffic-related fatalities. Our Waymo Driver powers a fully autonomous ride-hailing service and is adaptable to a diverse array of vehicle platforms and applications. With over ten million rider-only trips and experience from more than 100 million miles driven autonomously on public roads, Waymo is shaping the future of transportation.
Our simulation environment is among the most sophisticated ever created, utilizing deterministic logic, physical dynamics, and cutting-edge Generative AI to establish a training ground for the Waymo Driver. The Simulator Evaluation team tackles the challenging question: How can we mathematically validate that a virtual world is perceptibly 'real'?
We are on the lookout for a Staff Software Engineer who will take on the role of Technical Architect in this field. You will operate at the intersection of software engineering and AI, ensuring our simulated environments—whether governed by explicit rules or foundational models—accurately reflect reality.
In this Staff-level position, you will report directly to a Senior Staff Software Engineering Manager and function as a Technical Lead, connecting intricate technical metrics with overarching product strategies.
Your Responsibilities Will Include:
- Architecting Evaluation Standards: Define the 'Definition of Done' for simulation realism, anticipating product objectives (e.g., operational capabilities in snow or highway driving) and designing the evaluation roadmap to ensure our simulation fidelity evolves in alignment with onboard requirements.
- Acting as the System Critic: Create comprehensive mathematical frameworks to validate our hybrid virtual world, determining the balance between diverse evaluation needs—from verifying logical rules and dynamics to assessing the distribution quality of generative AI models.
- Building at Scale: Lead the development of large-scale, extensible evaluation platforms (in C++/Python), ensuring our metric pipelines are robust distributed systems capable of delivering clear, reproducible insights on petabytes of data.
- Providing Strategic, Cross-Functional Leadership: Serve as the technical liaison across various organizations, closely collaborating with AI research and other simulation teams. The evaluation workflows you design will facilitate rapid innovation and inform research trajectories.

