Hyperscale ML SoC Architect & Performance Modeling

Meta; Sunnyvale (CA); On-site; USD 300,000 - 540,000; Full time

Job summary

Meta is seeking a Machine Learning SoC Architect within the Silicon Engineering organization to design custom silicon powering Meta's AI and data-center workloads at scale.

You will own ASIC architecture specification, performance modeling, pre-silicon simulation, and microarchitectural exploration, partnering with RTL, silicon, and software teams to meet throughput, latency, and efficiency targets for hyperscale data centers.

Qualifications

• Bachelor's degree in Computer Science or Computer Engineering or equivalent.
• Experience with Computer Architecture concepts such as microprocessor architecture, memory systems, on-chip interconnection networks, hardware/software partitioning.
• 12+ years delivering high-performance ASICs with architecture focus.
• Experience in ASIC performance modeling or pre-silicon simulation.
• Proficiency in C++ and Python for simulation and tools.
• Experience with ML/data-center workloads on custom silicon.

Responsibilities

• Define and drive architectural definitions and performance analysis.
• Lead pre-silicon modeling and microarchitectural exploration of ASICs.
• Establish silicon roadmap strategy and performance validation.
• Partner with RTL and physical design teams for cross‑functional alignment.

Skills

Computer Architecture

C++ & Python

Performance modeling

Pre-silicon simulation

Cross-functional collaboration

Data-center/AI workload experience

Education

Bachelor's degree in CS/CE or equivalent

Master's or PhD in EE/CE or related field

Job description

Meta is seeking a Machine Learning SoC Architect within the Silicon Engineering organization to design custom silicon powering Meta's AI and data-center workloads at scale.

You will own ASIC architecture specification, performance modeling, pre-silicon simulation, and microarchitectural exploration, partnering with RTL, silicon, and software teams to meet throughput, latency, and efficiency targets for hyperscale data centers.