Sr. Technical Program Manager, Machine Learning Annapurna

Annapurna Labs is at the forefront of designing custom silicon and accelerator platforms that power AWS's AI and machine learning services. As compute density and power demands continue to grow, advanced cooling technologies are critical to unlocking the full performance potential of our Trainium and Graviton processors.

We are seeking an experienced Senior Technical Program Manager (TPM) to lead the AWS Advanced Cooling program. In this role, you will drive the evaluation, qualification, and deployment of next-generation cooling technologies — including microfluidics cold plates, jet impingement systems, pumped two-phase solutions, and advanced thermal interface materials — across Annapurna's custom silicon platforms.

You will serve as the central program leader connecting thermal engineering, reliability, manufacturing, supply chain, and datacenter operations teams to ensure advanced cooling innovations move from lab evaluation to production deployment at scale.


Key job responsibilities
- Program manage the end-to-end evaluation, qualification, and integration of advanced cooling technologies and chip packaging solutions for Annapurna's Trainium and Graviton server platforms.
- Partner with thermal/mechanical engineers, chip packaging engineers, reliability engineers, and silicon design teams to co-design cooling and packaging solutions from the earliest chip development stages.
- Drive chip packaging technology evaluation and integration, including advanced substrate designs, interposer technologies, 2.5D/3D packaging architectures, and thermal-aware packaging strategies that optimize both performance and cooling efficiency.
- Manage relationships with external cooling technology and chip packaging suppliers, driving standardized testing, performance benchmarking, and supply chain readiness assessments.
- Develop and maintain program roadmaps that align advanced cooling and chip packaging research timelines with new product introduction (NPI) schedules, ensuring innovations are production-ready when needed.
- Coordinate cross-functional teams spanning Hardware Engineering Services (HWEngS), Annapurna Labs, manufacturing, and datacenter operations to drive unified execution.
- Define and track key performance metrics — thermal performance, packaging reliability, manufacturing yield, cost, and sustainability — to inform technology selection decisions.
- Consolidate and manage program budgets across contributing organizations, ensuring efficient resource allocation.
- Communicate program status, risks, and recommendations to senior leadership, including Principal and Senior Principal engineers.
- Drive process improvements that enhance the team's operational effectiveness and scalability.
- Anticipate thermal and packaging requirements for future platforms across multi-year horizons, ensuring the technology pipeline stays ahead of silicon roadmaps.

About the team
The Accelerated and Trusted Security development team within AWS Hardware Engineering is responsible for delivering hardware as a service to our internal customers. We own all aspects of design, from concept to design to launch and then throughout the entire life of the product.