Software Engineer AI/ML, AWS Neuron Distributed Training Team

AWS Neuron is seeking a talented Software Engineer to join their Machine Learning Applications team, focusing on cloud-scale machine learning accelerators. This role is part of AWS's innovative Annapurna Labs division, which serves as the infrastructure backbone of AWS.

The position involves working with cutting-edge ML technologies, including large language models like GPT2 and GPT3, stable diffusion, and Vision Transformers. You'll be responsible for developing integrations with the Neuron SDK and major frameworks such as TensorFlow, PyTorch, and MXNet.

AWS offers a highly inclusive work environment with ten employee-led affinity groups and various innovative benefit offerings. The team strongly emphasizes work-life balance, believing that finding the right equilibrium between personal and professional life is crucial for long-term success and happiness.

The role provides excellent opportunities for growth and development, with a team structure that celebrates knowledge sharing and mentorship. Projects are assigned strategically to help team members develop into well-rounded professionals and take on increasingly complex challenges.

You'll be working with AWS Neuron, the complete software stack for AWS Inferentia and Trainium cloud-scale machine learning accelerators. This position offers the chance to work on significant projects that impact AWS's machine learning infrastructure and contribute to open-source projects.

The ideal candidate should have experience with ML infrastructure and systems, programming expertise in languages like C, C++, Java, or Perl, and familiarity with deep learning frameworks. The role offers competitive compensation based on location and experience, along with comprehensive benefits and potential equity compensation.

Join a team that's at the forefront of machine learning infrastructure, working on products that power AWS Nitro, ENA, EFA, Graviton, and F1 EC2 Instances, AWS Neuron, Inferentia and Trainium ML Accelerators, and scalable NVMe storage solutions.