Perception Machine Learning Engineer - Continuous Learning
Core
Build automated, closed-loop systems for continuous improvement of perception models in autonomous driving using active learning and data curation.
Role type
Senior IC perception machine learning engineer (continuous learning)
Builds
Large-scale data pipelines and active learning systems for the Waymo Driver
Domain
Autonomous driving / Computer Vision
Deliverable
production ML models
Required skills
active learning, large-scale ML data pipelines, data curation (balancing, subset selection, sampling), Python, deep learning frameworks (PyTorch or JAX), software engineering
Preferred skills
C++, building data-centric infrastructure from scratch, publications at top-tier conferences (CVPR, ICCV, ECCV, ICML, ICLR, NeurIPS, IROS, RSS, AAAI, IJCV, PAMI)
Technologies
PyTorch, JAX
Responsibilities
Design and scale data pipelines to mine and ingest massive volumes of sensor data; Deploy active learning algorithms to select impactful data for incremental model updates; Develop methods for evaluating real-world model performance and detecting regressions; Conduct large-scale experiments on data balancing and automated curation strategies; Develop pipelines for rare, safety-critical real-world scenarios; Utilize uncertainty estimation and embedding space analysis to guide data acquisition
Seniority
Senior, hands-on IC