Cloud Hardware Dev Engineer, Annapurna Labs - Server & Rack Delivery - Trainium

Do you want to build the backbone of Generative AI cloud at AWS? Do you want to build the future of the cloud for AI training and inference? Want to do industry leading work delivering continuous price performance improvements in the cloud for AI model training for multi billion variable LLMs? Come Join us in designing, delivering and operating AWS cloud offerings that enable high performance and scalability in AI/ML and HPC workloads.

The Annapurna Hardware Engineering team creates server designs for our own custom silicon products. Our designs are industry-leading in frugality and operational excellence, and are critical to the success of the AWS business and millions of customers who use AWS today. Our engineers solve challenging technology problems, and build architecturally sound, high-quality components to enable AWS to realize critical business strategies. What makes this role unique is your direct impact on AI-powered innovation - you'll be building intelligent systems that drive the debug and development of next-generation cloud technologies.

The ideal candidate for this role is a hardware engineer who will own the design, operations and strategy around keeping our hardware functional in the fleet. You are knowledgeable of the full technical stack - vertically from baremetal server hardware up to the software in userland, and everything in the middle. You have tremendous interest in cloud scale and curious how systems and software decisions impact the user. You insist on highest-standards and are able to develop tactical solutions/tools to diagnose and fix issues. You are an excellent systems debugger - finding interaction issues between components on server systems. You are a leader with strong organizational, planning, and communication skills. You are a builder!

Key job responsibilities

You will own the operational strategy and design of Annapurna"s Trainium servers to meet the high uptime requirements of our customers. You will be responsible for root causing complex problems in our fleet and driving scalable solutions. You will span hardware design, system integration and software tooling to simplify our fleet operations.

About the team

We are the System Integration team for Annapurna"s custom silicon accelerated servers. The team owns the entire system from the NPI phase and continues through launch and thereafter. Our software, services and automation are critical in design, EC2 instance build yields, launch go/no-go decisions, and fleet health thereafter. We build process automation systems to improve efficiency across the board.