Amazon
Software Development Engineer, Neuron Collectives, Annapurna Labs
$165K–$224K
Cupertino, CA · Full-time · Posted Oct 23, 2026
- Experience
- Experience in development in the last 3 years, or experience in embedded development in C/C++; no overall years requirement stated.
- Education
- Bachelor's degree in computer science or equivalent.
About the job
Optimizes Trainium collective communication software and topologies so AI training workloads can scale efficiently across AWS data centers.
What you'll do
- Develop and optimize C/C++ collective communication algorithms and training topologies.
- Analyze compute, bus bandwidth, processor, DMA, firmware, and workload metrics to identify and address bottlenecks.
- Collaborate with hardware teams to co-optimize software and Trainium silicon; build analysis frameworks and automation.
What they ask for
- Experience building complex software systems delivered to customers.
- Experience contributing to system architecture and design, including reliability and scaling.
- Bachelor's degree in computer science or equivalent.
- Knowledge of software/hardware/network development lifecycle practices; development experience in the last 3 years or embedded development in C/C++.
Nice to have
- Master's degree in computer science or equivalent.
- Experience with hardware/software integration and real-time systems; familiarity with collective communication algorithms or distributed training frameworks.
Pay
165,200.00 - 223,600.00 USD annually (base salary)
Tools
C/C++, Neuron Explorer, DMA, Trainium
AI focus
Optimizes collective operations and communication topologies for scaling AI training, including modern LLM training, across Trainium chips.
Model-written summary · the original posting is authoritative