Principal Engineer: AI Training & Inference on Trainium

Amazon Web Services (AWS); Seattle (WA); On-site; USD 200,000 - 271,000; Full time

Job summary

Amazon seeks a Principal Engineer to drive AI model performance across the lifecycle on AWS Trainium, with deep hands-on expertise in pre-training, post-training, and inference serving. You will write code, own critical paths, and influence architecture and practices across Annapurna Labs and partner teams, delivering scalable, cost‑efficient model deployment.

You will mentor engineers, collaborate with customers, and lead cross‑functional efforts to push the frontier of ML model performance on

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 optimization; and leadership/mentorship of engineers.

Responsibilities

• Drive AI model performance across pre-training, post-training, and inference serving on Trainium.
• Write code, own forward/backward paths, and make binding technical decisions.
• Influence across teams (EC2, EKS, VPC, Bedrock, HyperPod) and lead ML customers.
• Own model enablement from pre‑training to cost‑optimized serving at scale.
• Mentor senior engineers and raise the technical bar across the organization.

Skills

Distributed training

C/C++

Python

Technical leadership

Tools

JAX

PyTorch/XLA

TRL

veRL

NeMo-Aligner

TorchTitan/TorchAO

Job description

Amazon seeks a Principal Engineer to drive AI model performance across the lifecycle on AWS Trainium, with deep hands-on expertise in pre-training, post-training, and inference serving. You will write code, own critical paths, and influence architecture and practices across Annapurna Labs and partner teams, delivering scalable, cost‑efficient model deployment.

You will mentor engineers, collaborate with customers, and lead cross‑functional efforts to push the frontier of ML model performance on