AWS-Annapurna team develops the silicon used in our most advanced machine learning accelerator servers at cutting edge process nodes. These SoCs are used in massively scaled server clusters to provide best hardware platform for our customers to run training and inference workloads.
Our final product is a server, not just the silicon, so you will find yourself stretching beyond traditional silicon product engineering boundaries and dealing with various system issues and data sets, providing ample opportunities to learn.
We are looking for a Senior Technical Program Manager to own end-to-end execution for complex silicon products from development through qualification, production ramp, and high-volume manufacturing.
You will own the integrated product schedule across Product Engineering, Test Engineering, Design, DFT, System Validation, Packaging, Quality, Reliability, Supply Chain, Foundry, OSAT, and external silicon partners.
This role requires you to be the owner of the outcome, not the status update -
You will identify the critical path, expose technical and execution risks early, establish clear owners and dates, drive decisions, and hold teams accountable for closure. You will dive deep enough into technical issues to identify owners, distinguish real blockers from noise and escalate when a decision is required.
Success means products hit committed milestones with the required quality, capacity, and manufacturing readiness
Key job responsibilities
Own the end-to-end silicon product schedule from tapeout and first silicon through bring-up, characterization, qualification, production release, ramp, and HVM.
Build and maintain integrated execution plans across silicon, package, test, quality, reliability, manufacturing, and supply-chain dependencies.
Define milestones, entry and exit criteria, owners, critical paths, and recovery plans.
Drive Product Engineering, Test Engineering, Design, DFT, Packaging, System Validation, Quality, Reliability, and Supply Chain teams to committed deliverables.
Dive deep on execution blockers; identify owner, root cause, decision required, and recovery path.
Drive closure of risks affecting schedule, quality, yield, test coverage, qualification, cost, capacity, or production readiness.
Own execution/communication of schedules and needs across foundry, OSAT and other external manufacturing partners.
Identify long-lead dependencies early, including wafer starts, masks, test hardware, package material, sockets, load boards, qualification hardware, equipment capacity, and translate product demand into manufacturing-capacity requirements and ensure capacity is available when the product needs it.
Drive readiness reviews for first silicon, characterization, qualification, production release, and volume ramp.
Establish mechanisms that expose schedule health, risks, decisions, and dependencies without relying on manual status collection.
Escalate early when teams cannot meet commitments. Bring options, trade-offs, and a recommended path.
Drive decisions when engineering quality, schedule, cost, capacity, and product requirements conflict.
Communicate program health to senior leadership with clear facts: what changed, why it matters, what decision is required, who owns the action, and when it will close.
Identify recurring execution failures and drive mechanism changes so the same problem does not reoccur on the next product.
Scale common planning, readiness, and execution mechanisms across multiple silicon programs.
About the team
Our team is focused on delivering technology programs that create real impact for Amazon's customers. We value collaboration, clear communication, and a willingness to tackle hard problems together. We are building an inclusive environment where program managers and engineers work side by side to simplify complex systems and ship meaningful improvements. If you are looking for a role where your program management expertise directly shapes outcomes and where you can grow alongside talented colleagues, this is the team for you.
