Senior / Staff ML Training Optimization Engineer (Physical AI)

TL;DR

Senior / Staff ML Training Optimization Engineer (Physical AI): Building standardized distributed training frameworks for research and production with an accent on stability, efficiency, and performance profiling. Focus on designing CUDA kernels, implementing quantization-aware training, and optimizing resource usage for autonomous transportation models.

Location: Remote US & Canada / Hybrid in Dallas, TX, Phoenix, AZ, Pittsburgh, PA, San Francisco, CA, or Toronto, ON

Salary: $141,000 - $249,000

Company

hirify.global is a leader in Physical AI, developing technology for commercial autonomous trucks and robotaxis.

What you will do

• Build standardized distributed training frameworks for research and production to increase stability and efficiency.
• Profile model runtime and memory to identify and resolve performance bottlenecks.
• Evaluate and implement emerging technologies, including custom CUDA kernels and quantization-aware training.
• Collaborate with researchers and ML engineers to establish best practices for optimal resource usage.
• Develop tooling and dashboards to ensure the broad adoption of training and inference frameworks.

Requirements

• Degree in Computer Science, Robotics, or a similar technical field.
• Minimum of 4 years of industry experience.
• Proficiency in Python, C++, or Rust.
• Experience with deep learning frameworks such as PyTorch or Jax.
• Skill in profiling CPU and GPU code using tools like PyTorch Profiler and NVIDIA Nsight.
• Must be based in the US or Canada.

Nice to have

• Experience implementing custom CUDA kernels.
• Experience with Bazel in a monorepo environment.
• Experience with Kubernetes-based training platforms.

Culture & Benefits

• Competitive compensation and equity awards.
• Comprehensive medical, dental, and vision coverage.
• Unlimited vacation and flexible working hours.
• Work from home support.
• On-site perks: catered meals, gym, and games room.