ASIC Engineer, Architecture Kernel Development
Meta PlatformsJob Title
ASIC Engineer, Architecture Kernel Development
Role Summary
Join Meta's infrastructure organization as an ASIC engineer focused on architecture kernel development for AI accelerators. The role contributes to designing and validating compute kernels, performance analysis tools, and prototype software backends for machine learning accelerators used in Meta data centers.
Experience Level
Mid-level (typically 3–7 years); this posting requests 6+ years of relevant experience.
Responsibilities
Primary responsibilities center on implementing and validating high-performance AI kernels and the tooling needed to measure and improve their performance.
- Develop AI kernels for Meta AI models targeting Meta AI accelerators.
- Implement tools for static performance analysis and debug performance on dynamic performance simulators.
- Define compiler/ISA choices and microarchitectural optimizations to meet performance targets.
- Validate performance on RTL simulation and emulation platforms.
- Collaborate with AI/ML teams to prototype a PyTorch backend and optimize end-to-end kernel performance.
Requirements
Must-have technical skills and experience; preferred items listed separately.
- 6+ years mapping massively parallel workloads or kernel development for HPC or ML inference/compute.
- Proficient in C, C++, and Python.
- Experience with computer architecture and silicon development tooling (simulators, RTL validation, emulation).
- Proven ability to analyze and optimize performance across software, compiler, and microarchitecture layers.
Nice-to-have:
- Experience developing or optimizing compute/collective kernels with CUDA or SIMD models (SSE, AVX, NEON, Hexagon DSP, etc.).
- Understanding of GPU or AI-accelerator architectures and pre-silicon validation workflows.
- Experience with firmware, device drivers, API design, or library implementation.
Education Requirements
Minimum: Bachelor's degree in Computer Science, Computer Engineering or a relevant technical field, or equivalent practical experience. Preferred: Master’s or PhD in Electrical Engineering, Computer Engineering, or related areas for advanced ML performance optimization roles. "Equivalent practical experience" is explicitly accepted.
About the Company
Company: Meta Platforms
Headquarters: Menlo Park, California, United States
American technology company that develops social networking products (Facebook, Instagram, WhatsApp) and invests in virtual/augmented reality hardware and software through Reality Labs, focusing on connectivity, advertising, and immersive computing experiences.
