Waymo is an autonomous driving technology company with the mission to be the world's most trusted driver. Since its start as the Google Self-Driving Car Project in 2009, Waymo has focused on building the Waymo Driver—The World's Most Experienced Driver™—to improve access to mobility while saving thousands of lives now lost to traffic crashes. The Waymo Driver powers Waymo’s fully autonomous ride-hail service and can also be applied to a range of vehicle platforms and product use cases. The Waymo Driver has provided over ten million rider-only trips, enabled by its experience autonomously driving over 100 million miles on public roads and tens of billions in simulation across 15+ U.S. states.

We showed that we can deploy self-driving cars in the wild now we have to scale it. Scale is driven by large models and data: we are moving to ever larger models which generalize by being trained on more data. We need to optimize the model inference and training such that we can deploy ever larger models. Scale is driven by supporting multiple platforms: we are moving to new compute platforms, we need to support several onboard compute platforms and need to support offboard platforms e.g. for running simulations. We need optimizations which gracefully generalize to all platforms.

In this role you work embedded in an ML Engineering and Modeling team, you work hand-in-hand with the modeling team to drive scale and multi-platform support of the models. Optimizing neural model inference and training while moving to ever larger and ever more deeply integrated models (culminating in the end-to-end vision) makes this a field of technical growth and technical leadership opportunity. This role requires to follow the latest developments in efficient ML and bring those innovations to Waymo’s production systems.

You Will

• Optimize neural model architectures and systems for high performance across diverse GPU and TPU platforms (onboard and simulation).
• Enhance neural model and system performance for real-time constrained environments, such as Waymo's onboard systems.
• Develop and apply post-training and low-level optimizations (e.g., quantization, kernel optimization) to improve inference speed and reduce memory footprint on modern accelerators.
• Innovate new neural model architectures (e.g., sparse) and decoding strategies (e.g., speculative) to boost inference performance on GPUs and TPUs.
• Optimize training speed and efficiency for large, memory-bound models and I/O-bound fine-tuning processes.
• Foster collaboration with ML infrastructure, hardware, simulation, and Alphabet research teams.

You Have

• Education: Master’s degree or PhD in Computer Science, Engineering, or a related technical field.

• Experience:

• 6+ years in software development for neural model inference or training, with 3+ years specifically optimizing these on GPU/TPU architectures.

We Prefer

• Experience with ML-driven production systems, covering large-scale data, training, evaluation, and deployment.
• Proficiency in developing and optimizing large-scale vision, video, or multi-modal foundation models.
• Familiarity with end-to-end model development challenges.
• Ability to thrive in a fast-paced environment.

In this hybrid role you will report to the Engineering Director in Perception

Hybrid

The expected base salary range for this full-time position across US locations is listed below. Actual starting pay will be based on job-related factors, including exact work location, experience, relevant training and education, and skill level. Your recruiter can share more about the specific salary range for the role location or, if the role can be performed remote, the specific salary range for your preferred location, during the hiring process.

Waymo employees are also eligible to participate in Waymo’s discretionary annual bonus program, equity incentive plan, and generous Company benefits program, subject to eligibility requirements.

Salary Range

$281,000—$356,000 USD