AI Engineer, Data Scaling & Optimization

Avala; San Francisco (CA); Hybrid; USD 170,000 - 250,000; Full time

Benefits offered by this job

Equity

Health insurance

Dental insurance

Vision insurance

FSA

Unlimited PTO

Job summary

Avala is seeking a senior ML infrastructure engineer in San Francisco to scale neural network training infrastructure for Physical AI data needs. You will optimize data loading, pipelines, and distributed training across massive GPU clusters.

Join a team building cutting-edge autonomous systems data infrastructure, with equity and comprehensive health benefits. Competitive salary commensurate with experience and impact.

Qualifications

• 4+ years experience in ML infrastructure and training systems.
• Deep expertise in PyTorch or TensorFlow internals.
• Strong understanding of distributed computing and GPU optimization.
• Experience with large-scale data processing (petabyte+)
• Proficiency in Python and C++ for performance-critical code.
• Knowledge of modern ML architectures (transformers, vision models).
• Experience optimizing memory usage and training throughput.

Responsibilities

• Design and implement high-performance neural network training infrastructure
• Optimize training and evaluation code for maximum efficiency
• Build data pipelines handling petabytes of autonomous vehicle and robotics data
• Develop profiling and debugging tools for ML training systems
• Implement distributed training across large GPU clusters
• Work with research teams to enable rapid experimentation at scale

Skills

ML infrastructure

PyTorch internals

TensorFlow internals

Distributed computing

GPU optimization

Python

C++

Transformers

Vision models

Job description

Posted 10 months ago San Francisco, CA full-time

About This Role

Scale Avala's neural network training infrastructure to handle the massive data requirements of Physical AI. Optimize training pipelines, implement efficient data loading, and ensure our ML systems can leverage petabytes of multimodal robotics and AV data.

Key Responsibilities

• Design and implement high-performance neural network training infrastructure
• Optimize training and evaluation code for maximum efficiency
• Build data pipelines handling petabytes of autonomous vehicle and robotics data
• Develop profiling and debugging tools for ML training systems
• Implement distributed training across large GPU clusters
• Work with research teams to enable rapid experimentation at scale

Requirements

• 4+ years experience with ML infrastructure and training systems
• Deep expertise in PyTorch or TensorFlow internals
• Strong understanding of distributed computing and GPU optimization
• Experience with large-scale data processing (petabyte+)
• Proficiency in Python and C++ for performance-critical code
• Knowledge of modern ML architectures (transformers, vision models)
• Experience optimizing memory usage and training throughput

What We Offer

Competitive salary and meaningful equity in the team building Fleet Learning Infrastructure for Physical AI

Full health, dental, and vision coverage, plus FSA and unlimited PTO

Work on the closed-loop data infrastructure that the next generation of autonomous systems learns from

A team that built Tesla Autopilot's data infrastructure, now building something bigger