Sr. Machine Learning - Compiler Engineer III, AWS Neuron, Annapurna Labs

About the Role

Senior Software Engineer – Compiler, AWS Neuron. You will build the next‑generation Neuron compiler that transforms ML models from frameworks such as PyTorch, Tensor Flow, and JAX into forms optimized for AWS Inferentia and Trainium servers in Amazon’s cloud. You will solve complex compiler optimization problems to achieve optimum performance for a wide variety of ML model families, including large language models, vision transformers, and multi‑model systems.

Your work will involve technical communication with cross‑functional partners, pre‑silicon design, and contributing to open‑source communities.

• Design, implement, test, deploy, and maintain innovative software solutions that improve Neuron compiler performance, stability, and user interface.
• Collaborate with chip architects, runtime/OS engineers, scientists, and ML application teams to deploy ML models on AWS accelerators with optimal cost/performance.
• Lead efforts to build features that enhance the developer experience, including tool implementation and documentation.
• Provide expertise to our open‑source community collaborators (e.g., Stable
  
  HLO, OpenXLA, MLIR) and influence industry‑wide partners.
• Participate in design discussions, code review, and communication with internal and external stakeholders, including responding to compiler‑related questions in open forums.

In day‑to‑day work you will develop compiler optimization and verification passes, surface AWS accelerator features to developers, build tools for numerical error analysis, and debug compiler defects. You will engage in code reviews, design discussions, and liaise with both internal teams and open‑source communities. The environment is startup‑like, with focus on impactful outcomes.

• 5+ years of leading design or architecture of new and existing systems.
• 2+ years of experience developing compiler features and optimizations.
• Proficiency with at least one of the following languages: C++ (preferred), C, or Python.

• Master’s or PhD degree in computer science or equivalent.
• Experience with resource management, scheduling, code generation, and graph optimization.
• Experience optimizing Tensor Flow, PyTorch, or JAX deep‑learning models.
• Experience with multiple tool chains and instruction set architectures.

Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status.

Los Angeles County applicants:
This position requires safe, cooperative work with other employees, adherence to standards of excellence, effective communication, and compliance with all federal, state, and local laws and company policies. Criminal history may affect eligibility. Under the Los Angeles County Fair Chance Ordinance, qualified applicants with arrest and conviction records will be considered.

Base salary range:  –  USD annually (USA, CA, Cupertino).