The opportunity
Waymo is an autonomous driving technology company with the mission to be the world's most trusted driver. Since its start as the Google Self-Driving Car Project in 2009, Waymo has focused on building the Waymo Driver—The World's Most Experienced Driver™—to improve access to…
What you'll do
Architect Infrastructure: Design and scale the data pipelines needed to mine, ingest, and manage massive volumes of sensor data from our fleet.
Drive Model Improvement: Deploy active learning algorithms to continuously identify and select the most impactful data for training, ensuring our large models continuously adapt to new environments with incremental updates.
Ensure Model Quality: Develop methods and recipes for evaluating real-world performance of our models, and detecting regressions in model updates. Develop and maintain ground-truth free performance metrics.
Optimize Data Efficiency: Conduct large-scale experiments focused on data balancing, subset selection, and label quality optimization. Lead automated curation strategies—including smart pruning and downsampling—to minimize dataset bloat and maximize compute efficiency.
Solve Long-Tail Challenges: Develop robust mining, training and evaluation pipelines for rare, safety-critical real-world scenarios.
Innovate with Model Signals: Utilize uncertainty estimation, confidence scores, and embedding space analysis to uncover model blind spots and guide automated data acquisition.
What they're looking for
- Collaborate Cross-Functionally: Work closely with researchers and operations teams to iterate on the end-to-end model development lifecycle.
- A bachelor’s degree in Machine Learning, Robotics, or Computer Science. 3+: years of professional experience in Machine Learning and/or Computer Vision.
- Proven, hands-on experience applying active learning in production environments.
- Strong expertise in building large-scale ML data pipelines (mining, extraction, auto-labeling, ingestion).