Software Dev Engineer, Machine Learning Compilers

Overview In this role you will help build the compiler and software stack that enables large AI models to run on Amazon’s custom edge silicon. You’ll collaborate across hardware, software, and science teams to optimize deep learning workloads for constrained devices, including model quantization and backend code generation. You will influence the design of the compiler, performance tooling, and product-ready AI capabilities on consumer devices. This is a hands-on position with end-to-end ownership and a direct impact on edge AI experiences for customers. Compensation / Benefitshealth insurance (medical, dental, vision)401(k) matchingpaid time offparlental leaveRSUs and sign-on paymentsadoption of flexible benefits and mental health support ResponsibilitiesDesign and develop software stack for deep learning acceleratorDevelop compiler passes for graph ingestions, optimizations and partitioningDevelop backend code generation capabilities across heterogeneous platformsProfile, analyze and optimize system level performance, develop toolingParticipate in design reviews, API development, and documentationCollaborate with hardware, software, applied science and product teams to onboard more user experiences powered by Deep Learning acceleratorMentor and provide guidance to junior engineers Key requirements2+ years of non-internship design or architecture experience3+ years of non-internship professional software development experience3+ years of programming using a modern language such as Java, C++, or C# with OO design experienceExperience in developing and deploying LLMs in production on GPUs, Neuron, TPU or other AI acceleration hardwareteam collaborationmentorshipcommunicationembedded development in C/C++compiler development for accelerators or custom instruction setdesign patterns and reliability