Software Engineer, Systems ML — Compilers / Backend
Meta PlatformsJob Title
Software Engineer, Systems ML — Compilers / Backend
Role Summary
Join Reality Labs Silicon to design and implement a compiler toolchain that compiles PyTorch models down to binaries for custom on-device ML accelerators used in AR/VR and wearable AI devices. The role works with compiler, ML algorithms, firmware, and ASIC engineers to map ML graphs to hardware and optimize runtime performance and power.
Experience Level
Entry-level (2+ years of relevant experience required as stated in the posting).
Responsibilities
Key responsibilities include designing compiler passes, collaborating across hardware and algorithm teams, and optimizing toolchains for on-device ML inference.
- Design and implement compiler passes and optimizations for ML accelerators.
- Implement or enhance code generation targeting custom ML hardware.
- Work with algorithm and hardware teams to map ML graphs to hardware, model dataflows, and estimate power/performance trade-offs.
- Co-design hardware features to improve performance, power efficiency, and programmability.
- Develop and maintain ML libraries, intermediate representations, export formats, and analysis tools.
- Optimize kernels and compiled code to meet latency and inference targets.
- Perform design and code reviews; debug and resolve compiler and system issues.
- Collaborate with other compiler teams to evaluate and incorporate innovations.
Requirements
Must-have technical skills and experience; preferred items listed separately.
- Minimum 2+ years developing compilers, toolchains, runtimes, or code-optimization software.
- Proficient in Python and/or C/C++ for development, debugging, testing, and performance analysis.
- Experience in AI framework development or accelerating models on hardware architectures (GPU, TPU, or custom AI ASICs).
Nice-to-have:
- Experience with PyTorch, MLIR, TensorFlow, Caffe, or similar ML frameworks.
- Experience with machine-code generation or compiler back-ends for on-device inference.
- Contributed to compiler toolchains such as LLVM, MLIR, GCC, MSVC, or Glow.
- Experience developing high-performance kernels or runtime components and tuning for inference accelerators.
- Familiarity with deep learning algorithms (CNNs, RNNs) and responsible AI practices.
- Strong cross-functional communication and collaboration skills.
Education Requirements
Bachelor's degree in Computer Science, Computer Engineering, or a relevant technical field, or equivalent practical experience.
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.
