Sr. Power Engineer, Annapurna Labs, Machine Learning Hardware

About the role

Annapurna Labs is a division of AWS UC that designs silicon and software to accelerate innovation. This role involves designing, simulating, and validating power delivery solutions for machine learning products and ensuring these designs are reliable throughout the AWS fleet.

Responsibilities

• Drive improvements in availability and efficiency of power for AWS servers, including enhanced designs, testing, and manufacturing at suppliers.
• Identify and implement process improvements and automation in power design, validation, manufacturing, and testability.
• Develop comprehensive reliability models for server power using AWS production telemetry and component knowledge.
• Lead power technical aspects of high-demand machine learning rack-scale projects, including launching new services and features at scale.
• Collaborate with technology leaders to define and drive critical features and architectural advancements into rack-scale systems.
• Guide suppliers in meeting Amazon's power needs for future-generation products.
• Find simple solutions to complex problems and ensure the quality and reliability of the end-to-end system design.
• Engage in the operations of systems, identifying and resolving critical issues.
• Mentor team members and assist in their career development.
• Provide technical guidance to multiple teams to enhance their productivity and effectiveness.
• Advocate for the customer and ensure the best possible experience is delivered.

Requirements

• Bachelor's degree in Electrical Engineering, Computer Engineering, or a related field
• 8+ years of experience with hardware and software integration for embedded systems or hardware development
• 5+ years of experience managing Original Design Manufacturers (ODMs)
• 5+ years of experience as a lead technologist, architect, or engineer in the power space for a major technology, component, product, or product line
• Demonstrated expertise in enhancing power reliability and efficiency through design enhancements and failure mode analysis
• Experience with statistical analysis and modeling for reliability, performance, and/or cost

Qualifications

• Master's degree in Computer Science or a related field is preferred
• Experience with ML accelerators is preferred
• Experience with data center deployments of compute-related infrastructure is preferred