Senior Software Engineer, ML Evaluation Infra and Efficiency

The Waymo ML Frameworks & Efficiency team partners closely with Research and Production teams across Waymo to develop and deploy models in Perception and Planning that are core to our autonomous driving software. We help our partners by offering the best frameworks for the entire model development lifecycle, including training and evaluation. They are geared towards both scaling models with efficiency and solving problems unique to ML for autonomous driving. 

We are looking for engineers with ML system expertise to help us train, evaluate and improve pre-trained models to be deployed into Waymo Driver, and potential future products. You’ll work closely with researchers and modeling engineers across the company, and tackle the challenges of different evaluation scenarios for Waymo drivers, building large-scale evaluation platforms that can scale across compute, data, and environments to improve model intelligence and alignment with human drivers. 

You Will:

• Design and build distributed evaluation platforms for large-scale ML evaluation workloads.

• Profile evaluation platforms, identify performance bottlenecks (CPU, memory, I/O, network), and implement optimizations to improve inference speed and resource utilization.

• Collaborate with ML engineers to understand evaluation requirements and scenarios, and improve DevX of the evaluation infrastructure.

• Improve runtime goodput of ML inference workload and efficiency of metrics computations, ensuring scalability and reliability across distributed environments.

• Implement and maintain advanced ML infrastructure tools, including ML Pathways, JAX, Flume and TensorFlow.

You Have:

• B.S. in Computer Science, Math, or 3+ years equivalent real-world experience.

• Proficient in distributed systems design with an understanding of ML efficiency.

• Experience with ML frameworks, including TensorFlow, JAX, XLA.

• Solid programming skills in Python and C++.

• Practical familiarity with profiling tools to uncover performance bottlenecks.

We Prefer:

• MS in Computer Science, Math

• Familiarity with large-scale evaluation platforms for LLMs.