PreSilicon Power Modeling Leader - x86 SoCs

Advanced Micro Devices, Inc.
Austin, TX, United States
14 days ago

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Experience level
Expert
Compensation
$178,400.0
Working hours
Regular working hours

Tech stack

Artificial Intelligence BIOS Microarchitecture Python (Programming Language) Regression Analysis SystemC Grafana Caching Perf (Linux) Influxdb Physical Design Jenkins

Job description

AMD’s embedded and client x86 programs demand firsttimeright silicon. You will lead SoClevel presilicon power modeling-owning the methodology, models, and crossfunctional influence that translate into accurate Perf/Watt projections, robust DVFS/AVFS plans, and silicon correlation that derisks program power budgets., * Own the SoC power model endtoend using AMD’s converged methodology, delivering accurate workloadbased power estimates for CPU/GPU/NPU, fabric/interconnect, memory, and I/O.

  • Define usecase power envelopes across automotive, networking, storage, industrial, and robotics segments; partner with product/marketing to derive representative workloads and KPIs.
  • Drive DVFS/AVFS strategy (Pstates, Cstates, voltage/frequency points), telemetry requirements, and FW handoffs based on model outputs; provide early tables and guardrails to FW/BIOS.
  • Establish IP/subsystem power budgets; run sensitivity studies to guide architectural decisions (e.g., cache sizes, fabric frequencies, NPU/GPU configurations).
  • Build repeatable modeltosilicon correlation plans (CaC/static/dynamic splits, EDC/TDC interactions, droop mitigation) and maintain a power scorecard for program reviews.
  • Lead tool flows & automation (e.g., PowerArtist, PrimeTime PX; Pythonbased pipelines) to ingest activity (FSDB/VCD), generate reports, and publish dashboards for stakeholders.
  • Collaborate across Architecture, RTL/Design, Physical Design, FW/BIOS, and System PnP to ensure converged Perf/Watt outcomes from concept through bringup., AMD may use Artificial Intelligence to help screen, assess or select applicants for this position. AMD’s “Responsible AI Policy” is available here.

Requirements

  • Demonstrated experience building toplevel SoC power models with high correlation to silicon, including IPlevel VBM integration and workloadbased projections.
  • Handson with EDA/tooling: PowerArtist, PrimeTime PX (or equivalents), activity annotation (FSDB/VCD), UPF/CPF power intent, Python for automation.
  • Deep knowledge of x86 microarchitecture (CPU pipelines, caches), GPU/NPU basics, memory subsystems, interconnect/fabric behavior, and power delivery/PDN interactions.
  • Proven ability to define DVFS/AVFS strategies and drive FW telemetry requirements; familiarity with EDC/TDC/CaC semantics and their impact on power.
  • A validated SoC power model & scorecard that anchors program power budgets and executive reviews.
  • Workloadspecific DVFS/AVFS tables and telemetry specs for FW/BIOS enablement.
  • Correlation reports (estimated vs. measured) and recommendations that accelerate firsttimeright silicon. *, * Segment experience modeling power for automotive (AECQ100/ASIL) usecases, networking throughput workloads, storage I/O patterns, industrial/robotics duty cycles.

  • System performance modeling familiarity (GEM5/Simics/SystemC) to align Perf/Watt predictions and bottleneck analysis with architecture teams.
  • Experience creating PnP dashboards/regressions (Python/Jenkins/Grafana/InfluxDB) for continuous tracking., * B.S./M.S. in EE/CE/CS (or related) with strong industry experience in SoC presilicon power modeling/estimation and crossdomain collaboration.

About the company

At AMD, our mission is to build great products that accelerate next-generation computing experiences-from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you’ll discover the real differentiator is our culture. We push the limits of innovation to solve the world’s most important challenges-striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond. Together, we advance your career.

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on diversityjobs.com

Good distractions

Talks and stories from around this role — technically off-topic, practically not.

41 sec

Massive client data loss and bio-digital storage

Chris Heilmann +1 · LIVE

10:40 min

Visualizing Prometheus open metrics using custom Grafana dashboards

Stijn Polfliet · LIVE

2:42 min

Dissecting artificial intelligence layers from compute to applications

Christian Nagel Christian Nagel +3 · WWC Europe 2026

1:36 min

Accelerating time to market using virtual hardware platforms

Andrew Wafaa Andrew Wafaa · WWC 2024

1:04 min

Visualizing Keycloak performance via standard Grafana troubleshooting dashboards

Alexander Schwartz Alexander Schwartz · WWC 2025

4:47 min

Automating frontend performance metrics with Google Lighthouse

Miki Lombardi · JS Congress

Videos

See all

Related articles

See all