Sr. Pd Methodology Engineer, Annapurna Labs - Cloud Scale Machine Learning

Amazon Amazon · Big Tech · Austin, TX · Applied Science

This role focuses on the physical design methodology for ASIC chips used in AWS's machine learning servers, specifically AWS Inferentia and Trainium Systems. The engineer will define, develop, and optimize RTL2GDS flows, work with EDA tools, and improve cloud infrastructure utilization for physical design tasks. The role requires expertise in ASIC physical design, CAD flows, and scripting languages, with a focus on PPA and TAT improvements.

What you'd actually do

  1. Define, develop and deploy innovative physical design and verification methodologies (RTL2GDS) for ML Accelerator chips in advanced nodes
  2. Drive Optimizations in CAD flows/methodologies for PPA and TAT improvements
  3. Work with EDA tool vendors to evaluate new methods, resolve bugs, improve usability.
  4. Fine tune cloud infrastructure to improve compute and storage utilization for physical design work.
  5. Interface directly with RTL, Physical Design, Package Design, DFT teams to improve methodologies and efficiencies.

Skills

Required

  • BS + 10yrs or MS + 7yrs in EE/CS
  • 5+ years developing physical design methodology or CAD flows in synthesis, PNR, and sign-off areas for advanced technology nodes.
  • Proficient in programming/scripting languages (Perl, Python, C++)
  • Solid understanding of ASIC physical design, and methodologies including synthesis, place and route, STA, IR, formal and physical verification.
  • Demonstrated level of expertise in PD tools such as Innovus, ICC2, Fusion Compiler, STA, and Sign-Off.
  • Proven track record of delivering metric driven PPA flow development and support.

Nice to have

  • Expertise in high-performance, low-power physical design, and implementation techniques with industry standard synthesis, PnR, or Signoff tools.
  • Excellent programming skills in languages like Python, Perl, TCL, Shell, etc. Good understanding of algorithms with emphasis on optimizing algorithms.
  • Knowledge of technology nodes across foundries
  • Experience in evaluating multiple vendor solutions and driving tool decisions.
  • Knowledge of creating dashboards and status reports for various EDA tool outputs, QOR metrics and analyzing trends (synthesis, pnr, signoff etc)
  • Experience with machine learning
  • Excellent verbal and written communications
  • Ability to work in dynamic work environment with changing needs and requirements
  • Ability to provide mentorship, guidance to junior engineers and be a effective team player
  • Meets/exceeds Amazon’s leadership principles requirements for this role
  • Meets/exceeds Amazon’s functional/technical depth and complexity for this role

What the JD emphasized

  • advanced nodes
  • advanced technology nodes
  • advanced technology nodes