Software Development Engineer, HPC/ML Interconnect Engineer

We are seeking an experienced software engineer with low-level latency networking or interconnect expertise to optimize customer experience by designing systems that enable scaling network-intensive workloads over thousands of CPUs, GPUs, and TPUs. This role is on the forefront of AI/ML, focusing on optimizing networking for the latest AI workloads such as LLMs.

As part of the AWS Utility Computing (UC) organization, you'll support the development and management of various AWS services, including Compute, Database, Storage, IoT, Platform, and Productivity Apps. You'll work within Annapurna Labs, designing silicon and software that accelerates innovation for cloud solutions.

Key responsibilities:

• Design and optimize networking solutions for Machine Learning (ML) and High-Performance Computing (HPC) workloads on AWS
• Collaborate with cross-functional teams and engage with customers to gather feedback and improve offerings
• Develop low-latency networking and collective operations for HPC network fabric or machine learning accelerator cluster systems
• Troubleshoot complex networking issues and implement solutions at scale

Required skills:

• Extensive experience in low-latency networking and collective operations
• Proficiency in C/C++ and deep understanding of Linux and kernel-level programming
• Strong problem-solving skills and ability to troubleshoot complex networking issues
• Excellent communication skills for effective collaboration in a team environment

The role offers opportunities to work on cutting-edge AI/ML technologies, participate in innovative learning experiences, and benefit from a diverse and inclusive team culture. AWS values work-life balance and offers flexible working hours.

Join the Elastic Collectives team at Annapurna Labs and be part of shaping the future of networking solutions for ML and HPC workloads on AWS!