ML Inference & HW Codesign Engineer

Socket.dev; Mountain View (CA); On-site; USD 174,000 - 252,000; Full time

Job summary

Google DeepMind is building a next-generation compute platform for AI inference workloads, with emphasis on memory technologies, hardware acceleration, and SW/HW codesign.

We are seeking enthusiastic engineers to join our project, with a focus on ML compiler stack and inference infrastructure. This coding role requires initiative and collaboration across teams to push the boundaries of AI research and product development.

Qualifications

• Bachelor’s degree in CS, related field, or equivalent practical experience.
• Experience with Python and C++ software development.
• Experience with ML hardware accelerators, ML compiler stacks, and kernel development.

Responsibilities

• Direct full-stack Software role, focusing on ML compiler and inference infrastructure.
• Design and develop ML compiler stack to bridge AI workloads and low-level HW operators.
• Design and develop SW infrastructure to enable high performance inference.
• Drive SW/HW codesign, kernel optimization, performance tuning, and debugging.

Skills

Python

C++

ML hardware accelerators

ML compiler stacks

Education

Bachelor's degree in Computer Science or related field

Tools

JAX

PyTorch

XLA

Job description

Google DeepMind is building a next-generation compute platform for AI inference workloads, with emphasis on memory technologies, hardware acceleration, and SW/HW codesign.

We are seeking enthusiastic engineers to join our project, with a focus on ML compiler stack and inference infrastructure. This coding role requires initiative and collaboration across teams to push the boundaries of AI research and product development.