Software Development Engineer • Mid-Level

Who you are

You have a strong background in software development, particularly in optimizing large-scale systems for performance. Your expertise in CUDA and TensorRT allows you to drive inference optimization strategies effectively. You thrive in collaborative environments, working closely with scientists to influence model architectures and ensure optimal hardware utilization.

You possess a deep understanding of machine learning principles and have experience with robotics applications. Your ability to design and implement efficient compilation pipelines for complex transformer architectures sets you apart. You are also familiar with emerging optimization techniques, including ONNX Runtime and other ML compilers, which you leverage to achieve ambitious performance targets.

What you'll do

In this role, you will be instrumental in transforming novel research into high-performance production systems. You will collaborate directly with scientists to optimize large-scale transformer architectures specifically for robotics applications. Your responsibilities will include developing comprehensive benchmarking frameworks to measure and optimize model performance, as well as building robust monitoring solutions to ensure reliable model serving at scale.

You will explore and evaluate emerging optimization techniques, contributing to the overall strategy for inference optimization of large-scale foundation models. Your work will directly impact the performance and efficiency of Amazon's robotics initiatives, making you a key player in the next revolution in robotics.

What we offer

Amazon provides a dynamic work environment where innovation is encouraged. You will have access to cutting-edge technologies and the opportunity to work alongside some of the brightest minds in the field. The company offers competitive compensation, including equity and comprehensive benefits, ensuring that you are well-supported in your role. We encourage you to apply even if your experience doesn't match every requirement, as we value diverse perspectives and backgrounds.