ML Compiler Engineer — Build Next‑Gen DL Stack

Amazon Web Services (AWS); Cupertino (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Amazon Web Services (AWS) seeks skilled compiler engineers to join the Neuron Compiler team, building a state-of-the-art ML compiler stack for diverse domains from PyTorch, TensorFlow and JAX. You will collaborate with our accelerators like Inferentia/Trainium to optimize models and performance across systems.

You’ll design, develop and optimize compiler features, work with cross-functional teams, and contribute to hardware/software co-design across the backend stack including instruction

Qualifications

• Experience designing and optimizing compilers for ML workloads.
• Ability to work across software and hardware teams on compiler features.
• Strong coding skills in C/C++ and system-level programming.

Responsibilities

• Design, implement, test, deploy and maintain software to improve performance, durability, cost and security.
• Research implementations for best customer experiences and collaborate with Runtime, Frameworks and Hardware teams.
• Tackle challenging technical problems across stack layers and contribute to graph partitioning, code generation and ISA work.

Skills

Compiler engineering

ML compiler stack

C/C++

Tools

Inferentia

Trainium

Job description

Amazon Web Services (AWS) seeks skilled compiler engineers to join the Neuron Compiler team, building a state-of-the-art ML compiler stack for diverse domains from PyTorch, TensorFlow and JAX. You will collaborate with our accelerators like Inferentia/Trainium to optimize models and performance across systems.

You’ll design, develop and optimize compiler features, work with cross-functional teams, and contribute to hardware/software co-design across the backend stack including instruction