Compiler Architect, MTIA Software (Technical Leadership)

Job Title

Compiler Architect, MTIA Software (Technical Leadership)

Role Summary

Principal-level engineering role responsible for defining and executing the compiler architecture for Meta's MTIA (Meta Training and Inference Accelerator). The role focuses on compiler infrastructure, code generation, optimizations, and hardware-software co-design to maximize performance of ML workloads on custom accelerators.

Experience Level

Senior — 12+ years of software engineering experience with deep specialization in compiler development, code generation, or performance optimization for accelerators.

Responsibilities

Lead the technical direction and delivery of compiler infrastructure that maps ML workloads to custom accelerator hardware.

• Define and own the compiler architecture, roadmap, and technical strategy for MTIA (graph compilers, code generation, optimization).
• Design extensible compiler frameworks and intermediate representations to support evolving ML models.
• Develop and implement graph-level and low-level optimizations: operator fusion, memory planning, scheduling, and efficient mapping to hardware.
• Drive performance improvements across the compiler stack and remove bottlenecks from high-level graphs to low-level codegen.
• Collaborate with hardware teams on hardware-software co-design and influence accelerator architecture decisions.
• Establish validation practices for correctness, reliability, and performance at production scale.
• Integrate compiler infrastructure with ML frameworks (e.g., PyTorch) and collaborate with framework teams.
• Mentor engineers, lead architecture reviews, and communicate design and strategy to technical and non-technical stakeholders.

Requirements

Must-have technical skills and experience for the role.

• 12+ years of software engineering experience with deep specialization in compiler development, code generation, or performance optimization for accelerators.
• Proven experience architecting production compiler infrastructure for ML accelerators, GPUs, or custom silicon.
• Hands-on experience with compiler intermediate representations, optimization passes, and code generation techniques.
• Strong C++ experience building high-performance systems software and low-level optimization.
• Experience leading multi-year, cross-functional technical initiatives and driving execution across teams.
• Ability to influence technical direction and engineering practices via written proposals and design reviews.
• Nice-to-have: hardware-software co-design experience for ML accelerators or AI chips; experience with ML compiler stacks such as MLIR, XLA, TVM, Glow; familiarity with graph-level optimizations, operator fusion, memory planning, and scheduling; contributions to open-source, publications, or standards bodies; experience defining performance benchmarks and correctness validation; knowledge of ML model architectures (transformers, CNNs).

Education Requirements

Minimum: Bachelor's degree in Computer Science, Computer Engineering, or a related technical field, or equivalent practical experience. Preferred: Master's or PhD in Computer Science, Computer Engineering, or a related technical field. (Includes explicit allowance for 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.