Simulation Software Engineer for AI Accelerators

Meta; Menlo Park (CA); On-site; USD 154,000 - 217,000; Full time

Job summary

Meta seeks an experienced software engineer to join the MTIA software team in Menlo Park to work on simulators and early development platforms for pre-silicon AI accelerators. You will build robust simulation environments and collaborate across teams to drive performance and reliability.

The role emphasizes deep technical contributions, architecture awareness, and impact across Meta's AI infrastructure initiatives, with opportunities to influence hardware-software co-design.

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, or related field.
• Master's or PhD in Computer Science, Computer Engineering, or related field.
• 5+ years of experience in developing C++ codebase.
• 5+ years experience in developing Python codebase.
• Understanding of performance benchmarking for distributed at‑scale training and inference.

Responsibilities

• Contribute to developer infrastructure, including simulation and HW emulation platforms.
• Support networking and compute hardware acceleration techniques to improve ML inference and training.
• Implement simulation models for accelerator ASICs and evaluate data center performance.
• Perform architectural analysis to meet performance, scalability, and reliability requirements.
• Collaborate with architects and engineers to integrate simulation results into system design.
• Use instruction set simulators to define performant firmware for Meta's training/inference accelerators.
• Collaborate with hardware teams to ensure accurate modeling and simulation.

Skills

C++

Python

Distributed training

Performance benchmarking

SystemC/Qemu

Education

Bachelor's degree in CS/CE

Master's or PhD in CS/CE

Tools

SystemC

Qemu

Virtual Platform

CUDA

NCCL

PyTorch

Job description

Meta seeks an experienced software engineer to join the MTIA software team in Menlo Park to work on simulators and early development platforms for pre-silicon AI accelerators. You will build robust simulation environments and collaborate across teams to drive performance and reliability.

The role emphasizes deep technical contributions, architecture awareness, and impact across Meta's AI infrastructure initiatives, with opportunities to influence hardware-software co-design.