Machine Learning Engineer, Behaviour Models for Road Users at Zoox (Foster City)
Job Description
Zoox’s Offline Driving Intelligence team is hiring a machine learning engineer to build behaviour models of cars, cyclists and pedestrians that populate its simulations. The agents must look indistinguishable from real road users yet be fully controllable, so they can be prompted into rare, adversarial, safety-critical situations for testing. The ad asks for a Ph.D.
At a glance
- Team: Offline Driving Intelligence
- Location: Foster City, California
- Pay: $199,000 – $283,000 a year, as listed
- Apply: open until filled (re-checked 14 October 2026)
What you would do
- Develop imitation- and reinforcement-learning models that generate human-like driving plans for simulated agents
- Estimate the quality of plans for safety, progress, comfort and realism
- Build generative behaviour models (autoregressive or diffusion) that can be conditioned on scenario intent
- Build metrics and tools to analyse errors, and work with perception, planning, simulation and validation teams
Why this role matters
Testing a robotaxi on public roads alone would take far too long to meet every rare situation, so companies rely on simulation. The value of simulation depends on how realistic and controllable its virtual road users are, which is exactly what this team builds. Its models directly affect how fast Zoox can validate and ship its driving software.
What Zoox is looking for
- A Ph.D. in computer science, robotics or a related field
- Experience with imitation learning, RL or generative models
- Strong coding skills
Nice to have
- Experience with driving simulation or multi-agent behaviour
Pay and location
Zoox lists $199,000 to $283,000 a year for this role, plus other benefits.
How to apply
Apply through the official Zoox job posting. Zoox’s posting does not give a closing date, so the role is open until filled; ResearchJobs.in will re-check this listing on 14 October 2026. Check the posting for location and work-authorisation details before you apply.
See all our artificial intelligence jobs, or browse more research jobs on ResearchJobs.in.
Hiring institution: Zoox
Official advertisement: jobs.lever.co
How to prepare for this application
- Generative behaviour: show diffusion or autoregressive trajectory models.
- Controllability: conditioning models on intent is central here.
- Metrics: explain how you would measure realism.
- Ph.D. required: lead with your thesis.
About Zoox
Zoox, an Amazon subsidiary, designs purpose-built robotaxis and the self-driving software that runs them. Its Foster City, California teams work on perception, prediction, planning and simulation.
