data
Posted Dec 8, 2025Senior Data Scientist - Metrics
at May Mobility
Ann Arbor, United StatesOn-site
Responsibilities
- Develop and refine safety and comfort metrics for evaluating autonomous vehicle performance across real-world and simulation data.
- Build ML and non-ML models to detect unsafe, uncomfortable, or anomalous behaviors.
- Analyze large-scale drive logs and simulation datasets to identify patterns, regressions, and system gaps.
- Collaborate with perception, prediction, behavior, and simulation teams to integrate metrics into workflows.
Requirements
- Based in Ann Arbor, Michigan, May develops and deploys autonomous vehicles (AVs) powered by our innovative Multi-Policy Decision Making (MPDM) technology that literally reimagines the way AVs think.
- As we advance toward widespread deployment, the ability to measure safety and comfort objectively, accurately, and at scale is critical.
- Strong proficiency in Python, SQL, and data analysis tools (e.g., Pandas, NumPy, Spark).
- Strong understanding of vehicle dynamics, kinematics, agent interactions, and road/traffic elements.
- Excellent technical communication skills with the ability to clearly present complex model designs and results to both technical and non-technical stakeholders.
- B.S, M.S. or Ph.D. Degree in Engineering, Data Science, Computer Science, Math, or a related quantitative field. 5+ years of
- experience in data science, applied machine learning, robotics, or autonomous systems.
- experience developing or evaluating safety and/or comfort metrics for autonomous or robotic systems. Hands-on
- experience working with real-world driving logs and/or simulation data. Desired
- Background in motion planning, behavior prediction, or multi-agent interaction modeling. •
- experience doesn’t align perfectly with every qualification, we encourage you to apply anyway! You may be the perfect candidate for this or another role at May.
Experience
- 2+ years working in AV, ADAS, robotics, or another safety-critical domain involving vehicle behavior analysis. Demonstrated