← All roles

Software Engineer, AI Kernels & Performance Optimization — MTIA Software

Meta· Engineering
Apply on company site
Hybrid (Sunnyvale, CA)$184k - $257kFull-timePosted 2mo ago

About the role

Senior-level kernel and performance engineer within Meta’s MTIA Software team (AI Training and Inference Accelerator). Owns performance of MTIA workloads from inner kernel optimizations to end-to-end production readiness, collaborating with silicon teams on hardware/software co-design and delivering high-performance, extensible kernel frameworks for PyTorch operators across AI workloads.

Responsibilities

  • Design, implement, and optimize high-performance compute and communication kernels for MTIA accelerators, taking ownership from architectural analysis through production deployment
  • Profile and root-cause performance across the full stack — instruction scheduling, memory hierarchy and DMA behavior, on-chip interconnect, multi-device collectives — and drive fixes to the right layer (kernel, compiler, runtime, or hardware)
  • Build and extend kernel authoring frameworks, templates, and libraries so other engineers can reach high performance without deep architectural expertise
  • Deliver and maintain broad kernel coverage for PyTorch operators across recommendation, ranking, and generative AI workloads, in both eager and compiled execution paths
  • Partner with silicon architecture and design teams on hardware/software co-design: quantify the value of proposed features with real kernels, characterize rooflines pre-silicon, and advocate for needed changes
  • Work with compiler, runtime, framework, and product-facing teams to land end-to-end wins on production models
  • Investigate numerics and precision trade-offs, and design software mitigations that recover performance or accuracy lost to hardware limitations
  • Set technical direction for a kernel domain, write design documents, and mentor engineers on performance methodology and accelerator programming

Qualifications

  • Bachelor’s degree in Computer Science, Computer Engineering, related technical field, or equivalent practical experience
  • 6+ years of professional experience in high-performance computing, accelerator kernel development, compiler backends, or systems performance engineering
  • Proficiency in C++ and Python, including low-level systems programming, templates and generic programming, and performance-critical code
  • Demonstrated experience writing and optimizing kernels for parallel architectures (GPU CUDA, ROCm/HIP, SYCL/OpenCL, TPU or other AI ASICs, or SIMD/vector CPUs)
  • Working knowledge of computer architecture as it applies to performance (memory hierarchies, bandwidth, latency hiding, scheduling, vectorization, synchronization)
  • Rigorous, measurement-driven approach to performance (roofline or analytical models, profiling, explaining residual gaps)

Skills

Required:High-performance kernel design and optimizationPerformance profiling and debuggingKernel framework developmentPyTorch operator coverageHardware/software co-design collaborationNumerics and precision trade-offs

Benefits

  • Benefits information provided in the posting (link to benefits)
Last updated Oct 23, 2026. Interested? Apply on the company site.

About Meta

Shop the latest AI glasses and wearable technology from Meta.

Shop the latest AI glasses and wearable technology from Meta. Explore Ray-Ban Meta, Oakley Meta, display glasses, and cutting-edge VR headsets.

AI glassesOakley MetaRay-Ban MetaVR headsetsliquid-ai-recruiting-targetwearable technology
Company profileWebsite

More open roles at Meta