Principal and Senior Principal Engineer, AWS Compute and ML Services

We are hiring Principal Engineers and Senior Principal Engineers to join AWS Compute and ML Services.

AWS Compute and ML Services is the compute layer the internet runs on. Millions of customers — from early-stage startups to the world's largest enterprises — depend on us every second of every day. The problems here involve a level of scale that is unique among many axes: we enable a wide variety of hardware platforms, support dozens of geographies worldwide, land a large number of new servers every day, power a vast marketplace of AI models, and meet the security and commerce needs of the most diverse customer base in cloud computing.

We don't build on top of infrastructure. We are the infrastructure.

AWS Compute and ML Services is entering its most ambitious era. AI workloads are demanding unprecedented scale, interconnect bandwidth, and placement intelligence. Confidential computing is raising the bar on what "secure by default" means at the hardware level — we're building toward a world where zero-trust isn't a feature, it's the architecture. Graviton is redefining the price-performance curve with every generation.

And customers are pushing us to deliver bare-metal performance with full isolation, at global scale, with millisecond-level responsiveness.

We need engineers who can lead large teams to deliver what customers want in the AI era.

You won't be optimizing an existing system. You'll be deciding what the system is.

The Problems — How do you architect systems that serve more types of customers, across more hardware platforms, in more regions than anyone else — and make it feel seamless? How do you co-design silicon and software when the feedback loop is measured in years, not sprints? How do you build placement algorithms that simultaneously optimize availability, performance, cost, carbon, and constraints that haven't been invented yet?

How do you enable an AI model marketplace that scales from a single inference call to a 10,000-GPU training cluster? How do you make security guarantees that hold at every layer — from custom silicon to the application boundary — without sacrificing performance?

The Constraints — Microseconds matter. Billions of dollars of customer workloads depend on your decisions. There's no "move fast and break things" when you are the thing everything else is built on. You'll navigate complex tradeoffs across performance, security, and cost while maintaining the highest standards for operational excellence.

The Reward — Your work ships to every AWS customer. Not eventually. Not behind a feature flag. The abstractions you design become the foundation other services assume will always be there. You'll serve as the technical conscience for one of AWS's most critical platforms — the one everything else depends on.

• Own the technical architecture and hands-on delivery of critical systems — whether that's compute platforms, ML infrastructure, serverless runtimes, security foundations, or custom silicon integration
• Drive end-to-end system design decisions across the full stack, from hardware interfaces to customer-facing APIs, where latency, security, and reliability are non-negotiable
• Lead cross-team initiatives spanning multiple AWS organizations to deliver cohesive systems, not disconnected components
• Author and champion technical vision documents, influence product roadmaps, and represent AWS Compute and ML Services in executive-level architectural reviews
• Make hard architectural calls at the hardware-software boundary where tradeoffs are permanent and impact is measured in billions of customer-hours
• Mentor and develop senior engineers — raising the technical bar through design…