Senior Data Scientist

Zoox•Boston, MA

8d•$225,000 - $307,000

About The Position

Zoox is seeking a Senior Data Scientist to join its Perception Verification and Validation team. The team verifies and validates the perception stack of Zoox’s autonomous driving system by building metrics, datasets, and analyses that show how perception performance translates into driving behavior. This role owns evaluation strategy for major perception capabilities, partnering with perception and ML engineers, systems and safety engineers, and autonomy leadership to define what to measure, how to measure it rigorously, and how results inform release decisions. The work shapes how Zoox safely and efficiently ships perception improvements to its fleet by connecting model-level performance to end-to-end system behavior. The Senior Data Scientist will also mentor other data scientists and help set technical direction for perception validation.

Requirements

Master's degree with 7+ years of experience, or PhD with 3+ years of experience, in data science, applied science, or a related quantitative role such as Statistics, Mathematics, or Computer Science.
Expert proficiency with SQL and/or Spark (PySpark or Scala) and Python, with a track record of building production-quality analyses and pipelines on large-scale (terabyte-plus) distributed data.
Deep expertise in statistical modeling, inference, and decision making, including uncertainty quantification, and a history of connecting point and uncertainty estimates to high-stakes decisions.
Experience designing metrics and validating ML models, such as computer vision, sensor-based, or perception models, including discovering and measuring failure modes.
Demonstrated track record of leading technical projects across teams, mentoring other data scientists or engineers, and presenting results and recommendations to senior leadership.

Nice To Haves

Experience with autonomous driving, robotics, or other safety-critical systems, including system-level verification and validation or simulation.
Experience analyzing geospatial data and/or working with embeddings, such as similarity search or representation learning.
Skill in using AI tools to improve development efficiency and quality.

Responsibilities

Lead the design of metrics and evaluation frameworks for new perception capabilities, working with perception engineers early in model development to define ship readiness and its connection to end-to-end driving behavior.
Drive cross-team validation projects with perception, systems, safety, and autonomy stakeholders to establish metrics, evaluation datasets, and release criteria.
Develop statistical methods for uncertainty quantification, error estimation, and error propagation, including frequentist, bootstrap, and Bayesian approaches.
Build evaluation datasets and analyze large-scale geospatial fleet and simulation data to identify perception failure modes.
Translate findings into clear recommendations for engineering and executive leadership.
Mentor and guide data scientists by defining tasks, reviewing work, and raising the team's technical bar.