Principal Software Engineer, ML System Architect
Core
Architect and build a unified, large-scale AI platform leveraging Google DeepMind's foundation models (like Gemini) to accelerate the development of Waymo's autonomous driving software.
Role type
Principal Software Engineer, ML System Architect
Builds
A cohesive, efficient, and powerful offboard ML platform centered around a unified foundation model recipe for Waymo Driver.
Domain
Autonomous driving / Large-scale Machine Learning / Generative AI
Deliverable
production ML models
Required skills
Python, C++, JAX, TensorFlow, large-scale distributed training, TPU/GPU architecture, API design, data pipelines, tokenization strategies, model distillation
Preferred skills
multimodal and generative models, autonomous vehicle systems, robotics, simulation systems, open-source ML contributions
Technologies
Google DeepMind Gemini, JAX, TensorFlow, TPUs, GPUs
Responsibilities
Define technical roadmap for platform unification including codebase, data pipelines, and model architecture; Lead consolidation of forked foundation model codebases into a shared repository; Serve as primary technical interface with Google DeepMind to define integration APIs; Unify tokenization, data formats, and evaluation methodologies across use cases; Architect efficient large-scale distributed training and distillation setups; Provide technical mentorship and drive alignment across teams.
Seniority
Principal, strategy & mentorship