Staff Machine Learning Engineer, Tech Lead, Labeling Automation

Stack

Responsibilities

• ▹Lead the design and deployment of highly scalable auto-labeling pipelines to improve data throughput and reduce reliance on manual annotation
• ▹Develop automated anomaly detection and quality evaluation systems (auto-graders) to assess annotation accuracy and enforce quality standards
• ▹Train, optimize, and deploy advanced 2D and 3D computer vision models, from zero-shot models like SAM and real-time detectors like YOLO to bespoke 3D perception and tracking models
• ▹Fine-tune and deploy large vision-language models (VLMs) and LLMs, using prompt optimization and post-training techniques (SFT, RL) for complex labeling tasks
• ▹Set the long-term ML strategy, guide architectural decisions, and mentor senior and mid-level engineers
• ▹Collaborate closely with Perception, Planner, and Simulation teams to align labeling capabilities with evolving ML data needs

Requirements

• ▹8+ years of professional experience in software engineering and applied machine learning
• ▹Experience programming in C++ or Python
• ▹Experience building, evaluating, and deploying deep learning models for object detection, segmentation, and spatial tracking
• ▹Experience in large model training, distributed computing, and scaling deep learning architectures using frameworks like PyTorch or TensorFlow
• ▹Experience taking machine learning solutions through the entire lifecycle, from research and experimentation to robust, scaled production deployment

Nice to have

• ▹Experience building internal tooling for ML developers

Soft skills

What we offer

• ▹Discretionary annual bonus program
• ▹Equity incentive plan
• ▹Generous company benefits program (subject to eligibility)

About the company

Waymo is an autonomous driving technology company with the mission to be the world's most trusted driver. Building on the Google Self-Driving Car Project launched in 2009, the Waymo Driver has completed over 10 million rider-only trips and more than 100 million autonomous miles on public roads across 15+ US states, plus tens of billions of simulated miles.