ML Compiler Engineer for AI Accelerators

Amazon Web Services (AWS); Seattle (WA); On-site; USD 165,000 - 224,000; Full time

Job summary

Annapurna Labs (U.S.) Inc. is hiring a software engineer to join the AWS Neuron compiler team in Cupertino.

You will build the next generation Neuron compiler that translates ML models from frameworks like PyTorch, TensorFlow and JAX for AWS Inferentia and Trainium servers in the cloud. You will tackle compiler optimization challenges across Llama, Stable Diffusion, vision transformers and other models, partnering with customers and stakeholders while contributing to pre-silicon design and

Qualifications

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

Responsibilities

• Design, implement, test, deploy and maintain software solutions to improve Neuron compiler performance, stability and UI.
• Collaborate with chip architects, runtime/OS engineers, scientists and ML Apps teams to deploy models on AWS accelerators.
• Explore open-source tooling (StableHLO, OpenXLA, MLIR) to optimize ML workloads on AWS software/hardware.

Skills

Software development

C++/Java experience

Education

Master's degree or PhD in CS

Tools

OpenXLA

StableHLO

MLIR

PyTorch

JAX

Job description

Annapurna Labs (U.S.) Inc. is hiring a software engineer to join the AWS Neuron compiler team in Cupertino.

You will build the next generation Neuron compiler that translates ML models from frameworks like PyTorch, TensorFlow and JAX for AWS Inferentia and Trainium servers in the cloud. You will tackle compiler optimization challenges across Llama, Stable Diffusion, vision transformers and other models, partnering with customers and stakeholders while contributing to pre-silicon design and