{bc}
indeed

Compiler Engineer, Neuron Automated Reasoning Group

Amazon.com
Seattle, USA
Full-time
Onsite
USD 129000-148400 yearly
Discovered 1 months ago
Software programmingObject-oriented programmingC++JavaSystem design and architectureCompiler development

Full Job Posting

Role overview

Build the AWS Neuron compiler, which transforms machine-learning models from frameworks such as PyTorch, TensorFlow, and JAX for AWS Inferentia and Trainium servers.

Optimize compiler performance for varied workloads, including large language models, diffusion models, vision transformers, and multimodal models.

Responsibilities

  • Design, implement, test, deploy, and maintain software that improves compiler performance, stability, and user experience.
  • Create compiler optimization and verification passes, accelerator features, numerical-analysis tools, and defect fixes.
  • Collaborate with chip architects, runtime and operating-system engineers, scientists, machine-learning application teams, and stakeholders.
  • Help deploy customer machine-learning models on AWS accelerators with optimal cost and performance; contribute to pre-silicon design and product launches.
  • Participate in design discussions, code reviews, and open-source work.

Basic qualifications

  • At least 1 year of non-internship experience designing or architecting new and existing systems, including design patterns, reliability, and scaling.
  • Experience programming in at least one software programming language.

Preferred qualifications

  • Compiler experience or experience building machine-learning models with frameworks on accelerators such as GPUs is preferred, not required.
  • OpenXLA, StableHLO, or MLIR experience is a bonus.
  • Three or more years of full software development life-cycle experience is preferred.
  • A bachelor's degree in computer science or equivalent is preferred.

Product and team

AWS Neuron is an SDK for optimizing machine-learning workloads on AWS Inferentia and Trainium custom chips.

The team supports knowledge-sharing, mentorship, and engineering career growth.

Compensation

  • The posting lists annual starting pay of USD 129,000 for Seattle and USD 148,400 for Cupertino.

Apply for this job in 1 click

Skip the repetitive application forms

Install the Base Career Chrome Extension and autofill job applications across major job boards with your profile.

Sarah M.James T.Maya R.

Trusted by over 500,000 job seekers on Base Career

Start Free Today

More from this employer

More jobs at Amazon.com