Software Development Engineer, Infrastructure for Simulation and Science

Amazon.com Inc

MA

JOB DETAILS
SKILLS
Amazon Elastic Compute Cloud (EC2), Amazon Simple Storage Service (S3), Amazon Web Services (AWS), Architectural Services, Automation, Building Systems, Cloud Computing, Code Reviews, Data Management, Data Sets, Debugging Skills, GPU (Graphics Processing Unit), High Availability, Large-Scale Systems, Machine Learning, Machine Tool, Needs Assessment, On Call, Operational Audit, Operations Processes, Policy Evaluation, Problem Solving Skills, Reliability Engineering, Research & Development (R&D), Robotics, Robotics Software, Scalable System Development, Simulation, Software Development, Software Engineering, Team Player, Technical Writing, Training/Teaching
LOCATION
MA
POSTED
11 days ago

Are you excited about building the infrastructure that powers the next generation of robotics? The Infrastructure for Simulation and Science (ISS) team within Amazon Robotics is looking for a Software Development Engineer II to design and build scalable simulation platforms, ML training pipelines, and data infrastructure that accelerate robotics R&D.

You"ll work at the intersection of cloud infrastructure, simulation, and machine learning - building systems that enable science teams to train and evaluate robot policies at scale. Your work will directly support multiple robotics programs and have a multiplier effect across the organization.

Key job responsibilities

  • Design, build, and operate scalable simulation infrastructure on AWS (EKS, S3, EC2) that supports robotics R&D workflows.
  • Develop and maintain ML training pipelines using workflow orchestration tools (Metaflow, Ray, SkyPilot) for distributed compute.
  • Build data ingestion, streaming, and management systems for large-scale robotics datasets.
  • Improve platform reliability and reduce operational burden through automation, self-service tooling, and monitoring.
  • Collaborate with science teams to understand their infrastructure needs and translate them into reusable platform capabilities.
  • Participate in on-call rotations and operational reviews to maintain high availability of shared infrastructure.
  • Contribute to architectural decisions, code reviews, and technical documentation.

A day in the life

You might start the morning reviewing a deployment for a new data pipeline that streams teleoperation data into training workflows. After standup, you pair with a science team member to debug a distributed training job running on GPU clusters. In the afternoon, you work on automating a manual operational process that"s been eating into the team"s time, then wrap up with a code review for a teammate"s infrastructure improvement. Your customers are internal robotics science and software teams, and the problems you solve help them iterate faster on robot learning.

About the team

The ISS team provides the simulation and ML infrastructure backbone for Amazon Robotics R&D. We build the platforms that science teams use to train robot policies, run large-scale simulations, and bridge the gap between simulated and real-world environments. Our work spans cloud infrastructure (AWS, EKS, Kubernetes), simulation platforms (NVIDIA Isaac Sim), ML tooling (Metaflow, Ray, Weights & Biases), and data systems. We"re a collaborative, geographically distributed team that values ownership, operational excellence, and building things that scale. We care about reducing toil, enabling self-service, and making scientists more productive.

About the Company

A

Amazon.com Inc

At Amazon, we don’t wait for the next big idea to present itself. We envision the shape of impossible things and then we boldly make them reality. So far, this mindset has helped us achieve some incredible things. Let’s build new systems, challenge the status quo, and design the world we want to live in. We believe the work you do here will be the best work of your life.

Wherever you are in your career exploration, Amazon likely has an opportunity for you. Our research scientists and engineers shape the future of natural language understanding with Alexa. Fulfillment center associates around the globe send customer orders from our warehouses to doorsteps. Product managers set feature requirements, strategy, and marketing messages for brand new customer experiences. And as we grow, we’ll add jobs that haven’t been invented yet.

It’s Always Day 1
At Amazon, it’s always “Day 1.” Now, what does this mean and why does it matter? It means that our approach remains the same as it was on Amazon’s very first day – to make smart, fast decisions, stay nimble, invent, and stay focused on delighting our customers. In our 2016 shareholder letter, Amazon CEO Jeff Bezos shared his thoughts on how to keep up a Day 1 company mindset. “Staying in Day 1 requires you to experiment patiently, accept failures, plant seeds, protect saplings, and double down when you see customer delight,” he wrote. “A customer-obsessed culture best creates the conditions where all of that can happen.” You can read the full letter here

Our Leadership Principles
Our Leadership Principles help us keep a Day 1 mentality. They aren’t just a pretty inspirational wall hanging. Amazonians use them, every day, whether they’re discussing ideas for new projects, deciding on the best solution for a customer’s problem, or interviewing candidates. To read through our Leadership Principles from Customer Obsession to Bias for Action, visit https://www.amazon.jobs/principles
COMPANY SIZE
10,000 employees or more
INDUSTRY
Retail
FOUNDED
1994
WEBSITE
http://Amazon.com/militaryroles