The Simulator team builds state of the art simulations of realistic environments for the testing and training of the Waymo driver. We use machine learning extensively to model the real world, including but not limited to, realistic agents (vehicles, pedestrians, cyclists, motorcyclists etc.), roads, traffic control systems, weather etc. To increase the fidelity and steerability of the simulations, we employ large foundation models, trained on our massive datasets that allow us to quickly setup and rollout various scenarios to subject our driver to.

We are planning to set up a team in London UK to work closely with the teams in MTV and Oxford to build these foundation models out and to integrate them into various evaluation and training products. We are seeking research engineers to work on these exciting problems.

In this role you'll:

• Be part of a world class research engineering team to advance the state of the art of ultra realistic AV simulations using foundation models
• Collaborate with teams in Waymo Oxford to use large models to improve sim realism
• Design experiments that push the frontiers of AV simulations
• Develop metrics that measure the realism of simulated worlds
• Train and evaluate large models and integrate them into the simulator and its downstream applications
• Help hire outstanding research engineers from diverse backgrounds
• Be a part of a high performing, collaborative research engineering team that takes research ideas and productionizes them

It’s preferred if you have:

• Strong background and experience in applied Deep Learning
• Strong coding and design skills
• Experience in solving complex production problems using state of the art ML techniques
• Experience in taking research to production
• Expertise in Data Analysis or Data Science
• Experience in AV a plus