ML Research Scientist, Prediction and Smart Agents at Nuro (Mountain View)
Job Description
Nuro’s Prediction and Smart Agents team is hiring an ML research scientist to build state-of-the-art models that predict how surrounding traffic will behave. The models run onboard as part of the planning stack and offboard as realistic agents in closed-loop simulation, using approaches such as generative sequence modelling.
At a glance
- Team: Prediction and Smart Agents
- Location: Mountain View, California (headquarters)
- Pay: From $193,930 a year base, as listed
- Apply: open until filled (re-checked 14 October 2026)
What you would do
- Build models that predict the behaviour of other road users
- Deploy models onboard within the planning stack
- Create realistic smart agents for closed-loop simulation
- Explore new methods such as generative sequence modelling
Why this role matters
A self-driving car is only as safe as its guesses about what others will do. Good prediction models also make simulation realistic, which is how driving systems are tested before they reach roads. That double use gives this team’s work unusual leverage. It is also a good fit for researchers from trajectory forecasting, time-series or multi-agent learning backgrounds who want to see their models run on real vehicles.
What Nuro is looking for
- Research experience in machine learning, ideally in prediction or sequence modelling
- Strong implementation skills
- Ability to take research into production systems
Nice to have
- Experience with autonomous driving or multi-agent modelling
Pay and location
Nuro’s posting gives a base salary range starting at $193,930 a year, plus equity and benefits.
How to apply
Apply through the official Nuro job posting. Nuro’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.
How to prepare for this application
- Sequence models: show trajectory, time-series or generative sequence work.
- Multi-agent: interaction modelling is central to prediction.
- Simulation: experience with closed-loop evaluation is a plus.
- Production: models you deployed carry weight.
About Nuro
Nuro, founded in 2016, develops Level 4 self-driving technology, the Nuro Driver, which it licenses for robotaxis, delivery fleets and personal vehicles. It has raised more than $2 billion from investors including Uber, NVIDIA and Google.
