Neuron Compiler Engineer for ML Accelerators

Amazon Inc.; Cupertino, Northern (CA, KY); Hybrid; USD 129,000 - 148,000; Full time

Job summary

Amazon.com Services LLC is seeking a Compiler Engineer for the Neuron Automated Reasoning Group in Cupertino, CA. You will design and optimize the Neuron compiler to run ML models on AWS accelerators, collaborating with chip architects and ML teams to deliver high-performance software.

You will work with open-source ecosystems like StableHLO, OpenXLA and MLIR to push performance and usability, while exploring pre-silicon design and new features for Neuron compiler.

Qualifications

• 1+ years of non-internship design or architecture (design patterns, reliability and scaling) of new and existing systems experience.
• Experience programming with at least one software programming language.

Responsibilities

• Design, implement, test, deploy and maintain innovative software solutions to transform Neuron compiler's performance.
• Collaborate with chip architects, runtime/OS engineers, scientists and ML Apps teams to deploy models on AWS accelerators.
• Develop compiler optimization and verification passes and build features for AWS accelerators.
• Contribute to open-source projects like StableHLO, OpenXLA and MLIR to optimize ML workloads.

Skills

C++

Java

Education

Bachelor's degree in computer science or equivalent

Tools

OpenXLA

StableHLO

MLIR

Job description

Amazon.com Services LLC is seeking a Compiler Engineer for the Neuron Automated Reasoning Group in Cupertino, CA. You will design and optimize the Neuron compiler to run ML models on AWS accelerators, collaborating with chip architects and ML teams to deliver high-performance software.

You will work with open-source ecosystems like StableHLO, OpenXLA and MLIR to push performance and usability, while exploring pre-silicon design and new features for Neuron compiler.