Software Development Engineer, AWS Neuron

Requirements:

• At least 3 years of professional, non-internship software development experience.
• At least 2 years of non-internship experience designing or architecting new and existing systems, including design patterns, reliability, and scaling.
• Experience programming in at least one software language.
• Preferred: A masters degree or PhD in computer science or a related technical field.
• Preferred: At least 3 years writing production-quality code in an object-oriented language such as C++ or Java.
• Preferred: Experience with compiler design for CPUs, GPUs, vector engines, or ML accelerators.
• Preferred: Experience with open-source compiler tools such as LLVM or MLIR.
• Preferred: Familiarity with PyTorch, OpenXLA, StableHLO, JAX, TVM, deep-learning models, and algorithms.
• Preferred: Experience with modern build systems such as Bazel or CMake.
• Experience with compilers or building ML models using frameworks on accelerators such as GPUs is a plus, but not required.
• Familiarity with OpenXLA, StableHLO, or MLIR is an additional advantage.

Responsibilities:

• Design, build, test, deploy, and maintain software that improves the performance, stability, and usability of the Neuron compiler.
• Develop compiler optimization and verification passes, and create features that expose AWS accelerator capabilities to developers.
• Build tools for analyzing numerical errors and investigate and resolve compiler defects.
• Work with chip architects, runtime and operating-system engineers, scientists, and ML application teams to deploy customer models on AWS accelerators with strong cost and performance.
• Contribute to open-source projects and optimize advanced ML workloads across AWS software and hardware.
• Participate in pre-silicon design and help bring new products and features to market.
• Take part in design discussions and code reviews, and communicate with internal teams, external stakeholders, and open-source communities.
• Partner with customers and stakeholders to make the Neuron compiler performant and easy to use.

Technologies:

• AI
• AWS
• Bazel
• Cloud
• Hardware
• IoT
• Java
• LLVM
• PyTorch
• TensorFlow
• DevOps
• EC2
• GitHub
• Support
• Security

More

We are hiring a software engineer for the AWS Neuron Compiler team. Our vision is to make deep learning accessible to everyday developers and broaden access to AI hardware and software infrastructure. AWS Neuron is our SDK for optimizing complex ML models on AWS Inferentia and Trainium chips, which accelerate deep-learning workloads in the Amazon cloud. The role focuses on a next-generation compiler for models built with frameworks such as PyTorch, TensorFlow, and JAX, including large language models, Stable Diffusion, vision transformers, and multimodal models. Our wider AWS Utility Computing organization develops and manages services across compute, databases, storage, IoT, platforms, and productivity applications. We offer a collaborative, startup-like environment with knowledge-sharing, mentorship, thoughtful code reviews, career-development opportunities, and an emphasis on inclusion and work-life harmony. Compensation for the Cupertino, California position is $165,200–$223,600 annually, with sign-on payments and restricted stock units; our benefits include health coverage, 401(k) matching, paid time off, and parental leave.

last updated 43 week of 2026