Senior Data Scientist, Data Quality and AI Evaluation
Core
Define quality metrics and evaluation methods for auto-labeling pipelines; build human-verified golden datasets; analyze sensor and annotation data to identify root causes and systematic errors; develop automated approaches for failure clustering and data selection; build statistically sound audits and experiments.
Role type
Senior IC data scientist (autonomous driving data quality and AI evaluation)
Builds
Human-verified golden datasets, automated failure clustering tools, statistical audit frameworks, release acceptance criteria dashboards
Domain
Autonomous driving, computer vision, sensor data (camera, LiDAR), machine learning evaluation
Deliverable
production ML models
Required skills
Statistics, experimental design, sampling, hypothesis testing, uncertainty estimation, Python, SQL, precision/recall analysis, class imbalance handling, selection bias detection, data leakage detection, human-in-the-loop workflow design
Preferred skills
Autonomous driving domain knowledge, robotics, computer vision, 2D/3D annotations, multi-object tracking, occupancy networks, temporal/cross-sensor consistency analysis, annotator agreement measurement, versioned benchmark management
Technologies
Python, SQL
Responsibilities
Define quality metrics and evaluation methods for auto-labeling pipelines; Build and maintain human-verified golden datasets covering representative driving conditions; Analyze large-scale sensor and annotation data to identify root causes and systematic errors; Develop automated approaches for failure clustering and targeted data selection; Build statistically sound audits and experiments to establish release acceptance criteria; Partner with labeling operations to improve annotation guidelines and reviewer consistency