Sr. EMIR Engineer, Annapurna Labs - Cloud Scale Machine Learning

About the role

The Machine Learning Acceleration (MLA) team develops the Inferentia and Trainium SoCs, which power AI workloads in datacenters worldwide. We are seeking a Sr. EMIR Engineer with a strong track record of handling complex challenges.

Responsibilities

• Contribute to the EMIR sign-off of SoCs and maintain high quality standards.
• Work closely with architects, designers, verification engineers, power integrity engineers, and physical design experts.
• Define best practices in power grid design, develop test cases and vectors to model demand currents accurately, and model and sign-off EMIR considering board and package impact.
• Advance the state-of-the-art in EMIR analysis and modeling.
• Design and analyze power grids, setup flows for EMIR analysis, run EMIR analysis for large subsystems and SoCs.
• Partner with EDA tool vendors to develop and evaluate new methodologies for EMIR analysis.
• Develop and maintain dashboards for EMIR rollups.
• Collaborate with designers, architects, verification engineers, and physical design engineers to develop vectors for IR analysis, thermal analysis, and power estimation.
• Innovate on ways to model VR, board, and package impact on voltage drop more accurately.
• Participate in post-silicon power measurements and correlate with simulation.

Qualifications

• 5+ years of ASIC implementation, synthesis, STA, and physical design experience in deep sub-micron nodes (16nm or smaller).
• Expertise in EMIR analysis tools like Redhawk-SC or Voltus or similar.
• Deep circuit-level understanding of power, EM, and IR.
• Proficient in scripting languages (Tcl, Perl, Python, etc.).
• Good understanding of physical design, EM/IR, power integrity, and thermal at the die, package, board, and server levels.