Principal Engineer: AI Model Performance for Trainium/Neuron

Amazon; Cupertino (CA); On-site; USD 230,000 - 311,000; Full time

Benefits offered by this job

RSU program

Job summary

Amazon is seeking a Principal Engineer to lead AI model performance across the lifecycle on AWS Trainium, with deep hands‑on expertise. You will drive optimization from pre‑training through post‑training to inference serving, partnering with teams across EC2/EKS/VPC/Bedrock.

This is a deep, hands‑on builder role focused on scalable, cross‑functional impact. You will own the forward/backward path, influence architectural decisions, and mentor engineers while delivering production‑grade

Qualifications

• 10+ years of non‑internship software development experience
• 7+ years with distributed training or RL systems at scale, strong C/C++ and Python, experience with AI/ML model execution, inference serving, or compiler optimisation, and leadership/mentorship of engineering teams.

Responsibilities

• Drive AI model performance across pre‑training, post‑training, and inference serving on Trainium.
• Write code and own forward/backward paths; influence Neuron org and partner teams.
• Mentor senior engineers and raise technical standards across the organisation.
• Collaborate with EC2, EKS, VPC, Bedrock, HyperPod and customers to deploy scalable solutions.
• Shape hardware‑software co‑design discussions for future silicon.

Skills

C/C++

Python

Distributed training

RL systems

Tools

JAX

PyTorch/XLA

TRL

NeMo-Aligner

Job description

Amazon is seeking a Principal Engineer to lead AI model performance across the lifecycle on AWS Trainium, with deep hands‑on expertise. You will drive optimization from pre‑training through post‑training to inference serving, partnering with teams across EC2/EKS/VPC/Bedrock.

This is a deep, hands‑on builder role focused on scalable, cross‑functional impact. You will own the forward/backward path, influence architectural decisions, and mentor engineers while delivering production‑grade