Sr. Software Development Engineer, MLOPs

Description

We are looking for a Senior Software Development Engineer with deep expertise in machine learning operations to join the Data & Intelligence Foundation (DIF) team within Amazon. You will design, build, and operate the ML training infrastructure that enables robot learning at scale — from distributed GPU training pipelines to experiment tracking, data management, and model deployment.

We are looking for a Senior Software Development Engineer with deep expertise in machine learning operations to join the Data & Intelligence Foundation (DIF) team within Amazon. You will design, build, and operate the ML training infrastructure that enables robot learning at scale — from distributed GPU training pipelines to experiment tracking, data management, and model deployment. Our team is building the foundational ML platform that powers autonomous robotics across Amazon’s fulfillment network.

You’ll work at the intersection of large-scale distributed systems and cutting-edge ML research, turning novel vision-language-action models into production training workflows.

• Design and implement scalable ML training infrastructure on Kubernetes (EKS) with GPU scheduling and fault-tolerant distributed training
• Build and maintain CI/CD pipelines for ML models — from data ingestion through training, evaluation, and deployment
• Develop tooling for experiment tracking, hyperparameter optimization, and reproducibility
• Architect data pipelines that handle large-scale robotics datasets (telemetry, sensor recordings, demonstrations)
• Collaborate with research scientists to operationalize novel ML models into production
• Establish monitoring, alerting, and observability for training workloads and model performance
• Drive best practices for GPU fleet management, cost optimization, and capacity planning

You’ll spend your mornings reviewing training job health across our GPU cluster, debugging a distributed training run that hit a node failure overnight, and shipping a fix to our checkpoint recovery system. After lunch, you’ll pair with a research scientist to optimize their new imitation learning model for multi-node training, then architect a new data pipeline to ingest demonstration recordings from robot workcells.

You’ll close the day reviewing a PR from a teammate and planning the next iteration of our experiment tracking platform.

The Data & Intelligence Foundation (DIF) team builds the ML infrastructure platform for Amazon’s industrial robotics. We enable scientists to train, evaluate, and deploy models that power autonomous robots in fulfillment centers worldwide. We’re a small, high-impact team where every engineer shapes the architecture and directly accelerates robot intelligence. We value pragmatic engineering, deep technical ownership, and close collaboration with research.

• 5+ years of non-internship professional software development experience
• 5+ years of programming with at least one software programming language experience
• 5+ years of leading design or architecture (design patterns, reliability and scaling) of new and existing systems experience
• Experience as a mentor, tech lead or leading an engineering team

• 5+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience
• Bachelor s degree in computer science or equivalent
• Knowledge of Machine Learning and LLM fundamentals, including transformer architecture, training/inference life cycles, and optimization techniques
• Knowledge of ML frameworks including JAX, PyTorch, vLLM, SGLang, Dynamo, TorchXLA, and TensorRT

Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status.

Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit…