ML Compiler Engineer for AWS Neuron

Amazon Inc.; Cupertino (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Amazon is seeking a Software Development Engineer for the AWS Neuron compiler team at Annapurna Labs. You will design and implement compiler optimizations to transform ML models for AWS Inferentia and Trainium, collaborating with chip architects and ML teams to deliver high-performance solutions.

The role emphasizes experience in C++/Java, compiler design, and contributing to MLIR/OpenXLA ecosystems, with opportunities to engage in pre‑silicon design and new product features.

Qualifications

• 3+ years of non‑internship professional software development experience.
• 2+ years of non‑internship design or architecture experience in new or existing systems.
• Experience programming with at least one programming language.

Responsibilities

• Design, implement, test, deploy and maintain Neuron compiler improvements for performance and stability.
• Collaborate with chip architects, runtime/OS engineers, scientists and ML Apps teams.
• Contribute to open‑source toolsets (e.g., StableHLO, OpenXLA, MLIR) to optimize ML workloads on AWS accelerators.

Skills

Software development

Design architecture

C++/Java

Education

Master's or PhD in Computer Science

Tools

LLVM

MLIR

Job description

Amazon is seeking a Software Development Engineer for the AWS Neuron compiler team at Annapurna Labs. You will design and implement compiler optimizations to transform ML models for AWS Inferentia and Trainium, collaborating with chip architects and ML teams to deliver high-performance solutions.

The role emphasizes experience in C++/Java, compiler design, and contributing to MLIR/OpenXLA ecosystems, with opportunities to engage in pre‑silicon design and new product features.