Software Engineer - AI Training Systems (Trainium)

Amazon Inc.; Seattle (WA); On-site; USD 165,000 - 224,000; Full time

Job summary

Annapurna Labs, part of AWS, is seeking a Software Development Engineer to optimize collective operations for Trainium. You will work with the Neuron Collectives team to enhance communication patterns, parallel topologies, and DMA/firmware interfaces across data centers.

You will implement high-performance C/C++ code, analyze bottlenecks with Neuron Explorer, and partner with hardware engineers to maximize throughput while scaling AI training for modern LLMs.

Qualifications

• Experience building complex software systems delivered to customers.
• Experience contributing to architecture and design of new and current systems.
• Experience in development in the last 3 years, or embedded development in C/C++.

Responsibilities

• Enhance collective algorithms and topologies for optimal training performance
• Use tools like Neuron Explorer to identify bottlenecks in compute and bus bandwidth utilization
• Monitor and analyze processor, DMA, firmware, and workload metrics
• Optimize collective operations to scale AI compute across the data center
• Work closely with the hardware team to co‑optimize software and Trainium silicon
• Develop and optimize C/C++ implementations of collective communication patterns
• Investigate and implement improvements for specific training topologies used by modern LLMs
• Build and maintain analysis frameworks and automation solutions

Skills

C/C++ embedded development

Software architecture & design

Complex software systems

Education

Bachelor's degree in computer science or equivalent

Job description

Annapurna Labs, part of AWS, is seeking a Software Development Engineer to optimize collective operations for Trainium. You will work with the Neuron Collectives team to enhance communication patterns, parallel topologies, and DMA/firmware interfaces across data centers.

You will implement high-performance C/C++ code, analyze bottlenecks with Neuron Explorer, and partner with hardware engineers to maximize throughput while scaling AI training for modern LLMs.