Waymo is seeking to optimize and enhance the performance of machine learning workloads across both cloud infrastructure and autonomous vehicles, focusing on efficient deep learning model deployment, model compression, and system-level performance improvements to contribute to the safety and reliability of their autonomous driving systems.
Requirements
- At least 5 years of experience developing solutions within ML systems or ML software stacks (e.g., PyTorch, JAX, TensorFlow, runtime libraries, ML compilers)
- Deep understanding of ML system architecture, performance analysis, and optimization tools
- Proficiency in Python and C++ programming languages
- Familiarity with hardware architectures of ML accelerators such as GPUs and TPUs
- Knowledge of model optimization techniques and efficient deep learning methods, including foundation models and large language models (LLMs)
- Experience with GPU and TPU hardware, along with related system software
- Preferred qualifications include a PhD in relevant fields, experience with ML hardware accelerators, and advanced knowledge of model compression and optimization techniques.
Responsibilities
- Analyze ML system workloads deployed on cloud infrastructure and autonomous vehicles to identify bottlenecks and optimize performance
- Implement model compression, efficient deep learning techniques, and system software improvements to enhance ML workload efficiency
- Develop and maintain tools and frameworks to support the lifecycle of machine learning workflows, including feature management, experiment tracking, and model deployment
- Conduct performance analysis and tuning of ML models and systems, ensuring optimal resource utilization and speed
- Stay abreast of advancements in ML hardware and software, integrating innovative solutions into Waymo’s systems
- Lead collaboration efforts with Waymo’s ML scientists across perception, planning, research, and simulation teams to identify performance improvement opportunities
- Work at the intersection of hardware and software, focusing on efficient deep learning model deployment, model compression, and system-level performance improvements.
Other
- Minimum of 2 years of experience as a technical lead, including project planning, stakeholder engagement, mentoring, and goal management
- Manage and drive projects from initial proposal through execution, including mentoring junior engineers and ensuring timely delivery
- Reporting directly to the Senior Manager of Runtime and Optimization
- This hybrid role offers a unique opportunity to influence the entire ML stack, from model development to deployment on resource-constrained hardware, making a tangible impact on the future of autonomous mobility.
- Inclusive and innovative work environment supporting diversity and collaboration
