Software Development Engineer, Neuron Collectives

Salary: $165,200 - 223,600 per year Requirements:

• We need experience building complex software systems that have been successfully delivered to customers
• We need experience contributing to the architecture and design of new and current systems, including architecture, design patterns, reliability, and scaling
• We require a bachelors degree in computer science or an equivalent background
• We need knowledge of engineering practices and patterns across the full software, hardware, and networks development life cycle, including coding standards, code reviews, source control management, build processes, testing, certification, and live site operations
• We need recent development experience within the last 3 years, or embedded development experience in C/C++
• A masters degree in computer science or an equivalent background is preferred
• Experience with hardware/software integration and real-time systems is preferred
• Familiarity with collective communication algorithms such as all-reduce and all-gather, or with distributed training frameworks, is preferred

• We enhance collective algorithms and topologies to improve training performance
• We use tools such as Neuron Explorer to uncover bottlenecks in compute and bus bandwidth utilization
• We monitor and evaluate processor, DMA, firmware, and workload metrics
• We optimize collective operations to scale AI compute across the data center
• We work closely with the hardware team to co-optimize software and Trainium silicon
• We develop and optimize C/C++ implementations of collective communication patterns
• We investigate and implement improvements for training topologies used by modern LLMs
• We build and sustain analysis frameworks and automation solutions

• AI
• AWS
• EC2
• Embedded
• Firmware
• Hardware
• Support
• Model Training
• Cloud
• Flow
• Machine Learning

More:

We are Annapurna Labs, part of AWS, and we design software, systems, and chips that help power EC2 infrastructure and improve the AWS customer experience. Our Neuron Collectives team owns the software stack behind collective operations for Trainium, the purpose-built AI training chip used to scale frontier model training across thousands of chips in the data center. This role gives you the chance to work at the intersection of hardware, firmware, and distributed systems, alongside a collaborative team that values inclusion, mentorship, learning, and work-life balance. We offer flexible working hours, strong career growth support, and comprehensive benefits that include health coverage, 401(k) matching, paid time off, parental leave, and additional compensation components such as sign-on payments and restricted stock units. The role is based in Cupertino, California, with a listed annual base salary range of USD 165,200 to 223,600.

last updated 37 week of 2026