Machine Learning Compiler Engineer, AWS Neuron

Salary: $165,600 - 165,600 per year Requirements: We require 3+ years of non-internship professional software development experience. We require 2+ years of non-internship experience in design or architecture for new and existing systems, including design patterns, reliability, and scaling. We require experience programming in at least one software language. We prefer a Masters degree or PhD in Computer Science, or a related technical field. We prefer 3+ years of production-grade development experience in object-oriented languages such as C++ or Java. We prefer experience with compiler design for CPU, GPU, vector engines, or ML accelerators. We prefer experience with open-source compiler toolchains such as LLVM or MLIR. We prefer experience with technologies such as PyTorch, OpenXLA, StableHLO, JAX, TVM, deep learning models, and algorithms. We prefer experience with modern build systems such as Bazel or CMake. Responsibilities: We will design, implement, test, deploy, and maintain software solutions that improve Neuron compiler performance, stability, and user experience. We will work closely with chip architects, runtime and OS engineers, scientists, and ML applications teams to deploy state-of-the-art ML models on AWS accelerators with strong cost/performance outcomes. We will build and work with open-source technologies such as StableHLO, OpenXLA, and MLIR to optimize advanced ML workloads on AWS software and hardware. We will create compiler optimization and verification passes. We will build features that expose AWS accelerator capabilities and nuances to developers. We will develop tools to analyze numerical errors. We will investigate and resolve the root causes of compiler defects. We will participate in design discussions, code reviews, and communication with internal and external stakeholders, including open-source communities. Technologies: AWS Bazel Cloud Hardware Support Java LLVM Machine Learning PyTorch TensorFlow AI EC2 IoT Security More: We are AWS Neuron, the SDK that optimizes complex machine learning models on AWS Inferentia and Trainium, our custom chips built to accelerate deep-learning workloads. This software engineer role sits on the Compiler team, where we are building the next generation Neuron compiler for models written in frameworks such as PyTorch, TensorFlow, and JAX and deployed on AWS cloud servers. You will help us tackle challenging compiler optimization problems for large language models, stable diffusion, vision transformers, and multimodal systems while partnering across teams in a startup-like environment. We support a broad mix of experience levels, value mentorship and knowledge sharing, and offer a culture that emphasizes inclusion, work-life balance, career growth, and flexibility. Amazon also provides comprehensive benefits including health coverage, 401(k) matching, paid time off, parental leave, sign-on payments, and RSUs, with the listed base salary range for Cupertino, CA at 165,600.00 USD annually. last updated 37 week of 2026