Meta
ASIC Engineer, Performance Architecture and Modeling
$178K–$250K
Sunnyvale, CA +2 · Remote · Full-time · Posted Jul 6, 2026
- Location
- Sunnyvale, CA
Austin, TX
Remote, US - Experience
- 8+ years in ASIC design, silicon engineering, or a related technical field; 5+ years in ASIC performance modeling, microarchitectural analysis, or pre-silicon simulation for custom silicon or SoC designs.
- Education
- Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience.
About the job
Defines performance architecture and pre-silicon modeling for custom ASICs supporting Meta’s data center and AI workloads.
What you'll do
- Develop performance modeling methods and simulation environments, and analyze microarchitectural trade-offs across compute, memory, interconnect, and I/O.
- Build and validate performance models and workloads, and investigate differences between modeled and measured silicon performance.
- Partner with architecture, design, software, and systems teams to shape specifications, optimize performance, and communicate recommendations.
What they ask for
- Bachelor's degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
- Proficiency in C++ and Python.
- Experience building cycle-accurate or transaction-level performance models using C++ and SystemC for complex digital systems.
- Experience with performance analysis of data center, AI accelerator, or high-performance computing workloads on custom silicon.
Nice to have
- Familiarity with post-silicon performance validation and model-to-hardware correlation.
- Experience with high-level synthesis, power-performance-area trade-off analysis, or PPA-driven microarchitectural optimization.
Pay
$178,000/year to $250,000/year USD base salary; bonus and equity amounts not stated
Tools
C++, Python, SystemC, SystemVerilog, VHDL
AI focus
Develops and models ASIC performance for AI workloads, including machine learning training and inference applications.
Model-written summary · the original posting is authoritative