ML Infra Engineer: Autonomy Optimization & Compiler

Nuro; Mountain View (WY); On-site; USD 160,000 - 241,000; Full time

Benefits offered by this job

Annual bonus

Equity

Benefits

Job summary

Nuro is seeking an Autonomy ML Infrastructure engineer to design, build, and deploy core ML infrastructure for the autonomy stack. You will optimize models, validate deployments, and work with engineers across perception, mapping, and behavior to deliver end-to-end learned ML solutions.

The role emphasizes quantization, pruning, and scalable model tooling, with a base pay range disclosed and bonus/equity opportunities as part of total compensation.

Qualifications

• 2+ years of relevant experience in ML optimization infrastructure.
• Experience with ML optimization techniques such as quantization and pruning, and ML compilers.
• Experience maintaining, profiling, and optimizing GPU ML compilers & runtimes.
• Proficient in Python and working experience with C++ and CUDA.
• Working experience deep learning frameworks (like PyTorch, Jax, Tensorflow, Keras).

Responsibilities

• Optimize Nuro’s autonomy stack with cutting-edge optimization techniques like quantization, low precision inference, and model pruning.
• Work with autonomy engineers to optimize, validate, and deploy large language models.
• Develop and maintain a world-class model compiler framework, FTL.
• Write robust, high-quality software to increase our confidence in our vehicle’s ability to navigate safely on-road.
• Collaborate with ML domain experts and engineers across behavior, perception and mapping to design end-to-end learned ML solutions.

Skills

ML optimization

Quantization

Pruning

ML compilers

Python

C++

CUDA

PyTorch

Tools

FTL compiler

Job description

Nuro is seeking an Autonomy ML Infrastructure engineer to design, build, and deploy core ML infrastructure for the autonomy stack. You will optimize models, validate deployments, and work with engineers across perception, mapping, and behavior to deliver end-to-end learned ML solutions.

The role emphasizes quantization, pruning, and scalable model tooling, with a base pay range disclosed and bonus/equity opportunities as part of total compensation.