Startups · AI

Manager - Model Evaluation

Zoox · Foster City, CA · Hybrid

← All jobs
About Zoox

Zoox is a purpose-built autonomous vehicle designed for riders, not drivers. Learn more about the Zoox robotaxi and the future of ride-hailing. Backed by Lux.

About the role

As the Manager of Model Validation & Verification (VnV) for Behavior Autonomy, you will lead an engineering and data science team responsible for evaluating, benchmarking, and validating the machine learning models and behavioral algorithms that drive our autonomous vehicle (AV) prediction and planning stacks.

What they're looking for

  • Domain Knowledge: Strong background in ML model validation, behavioral evaluation frameworks, system-level performance benchmarking, and statistics
  • Technical Depth: Proven familiarity with modern C++/Python ML environments, simulation frameworks, high-throughput ML evaluation pipelines
  • Cross-Functional Leadership: Demonstrated ability to navigate complex organizational trade offs between release velocity, compute cost, and safety rigor
More about this role

As the Manager of Model Validation & Verification (VnV) for Behavior Autonomy, you will lead an engineering and data science team responsible for evaluating, benchmarking, and validating the machine learning models and behavioral algorithms that drive our autonomous vehicle (AV) prediction and planning stacks.

You will own the statistical frameworks, offline/online evaluation metrics, and validation pipelines that ensure our behavioral models operate safely, comfortably, and predictably. Operating at the intersection of Data Science, Machine Learning, and Safety Engineering, you will partner closely with Autonomy Software, Prediction, Planner, and ML Operations teams to establish data driven release criteria for our AV fleet.

  • Team Leadership & Execution: Lead, mentor, and scale a high-performing team of Data Scientists, ML Validation Engineers, and Software Engineers while driving roadmaps, sprint execution, resource allocation, and high-throughput model releases with rigorous safety guardrails. Culture of Rigor: Foster a culture of statistical excellence, healthy skepticism, proactive risk tracking, and data-driven decision-making.
  • Validation Strategy & Methodologies: Define...

Read the full posting on Zoox's site ↗

Autonomy Software

Build your edge while you search

Free tools for founders and investors, plus VC Unfiltered, our take on startups, venture and the people who build them.