Senior ML Runtime Engineer — Real-Time Inference

Waymo; California (MO); Hybrid; USD 213,000 - 263,000; Full time

Benefits offered by this job

Bonus program

Equity incentive plan

Benefits

Job summary

Waymo is seeking an engineer to build the next generation ML onboard inference runtime and related deployment tooling, spanning from efficient deep learning models to optimized serving. You will work across the ML stack for both onboard compute and large-scale data center environments.

The role combines systems engineering with ML compiler and runtime work, requiring collaboration with perception, planner and research teams, and offers opportunities to influence performance across Waymo's

Qualifications

• BS/MS in CS/EE/Deep Learning or related field.
• 5+ years professional software engineering in ML systems.
• 5+ years production C++ experience.
• 3+ years Python and major deep learning frameworks (PyTorch, JAX).
• Experience optimizing ML software for GPUs/TPUs and hardware accelerators.
• Experience building low-latency, highly concurrent distributed backends.

Responsibilities

• Architect and develop efficient ML runtime and serving system for onboard and data-center environments.
• Lead integration of ML inference runtimes across onboard and offboard domains with real-time constraints.
• Drive migration toward a JAX-native runtime (OpenXLA/PjRT, TensorRT).
• Collaborate with Waymo ML teams to analyze workloads and apply hardware-aware optimizations.
• Design tooling for profiling, benchmarking, and identifying end-to-end bottlenecks.

Skills

C++

Python

Deep Learning

Distributed systems

Education

B.S./M.S. in CS/EE/Deep Learning

Tools

OpenXLA/PjRT

TensorRT

ONNX Runtime

TVM

CUDA

JAX

PyTorch

Job description

Waymo is seeking an engineer to build the next generation ML onboard inference runtime and related deployment tooling, spanning from efficient deep learning models to optimized serving. You will work across the ML stack for both onboard compute and large-scale data center environments.

The role combines systems engineering with ML compiler and runtime work, requiring collaboration with perception, planner and research teams, and offers opportunities to influence performance across Waymo's