Compiler Engineer (PyTorch / Triton / LLVM / MLIR)

Majestic Labs ai; Los Altos (CA); On-site; USD 120,000 - 170,000; Full time

Job summary

Majestic Labs ai, based in Los Altos, California, is looking for an experienced Senior Compiler Engineer to join their dynamic team. You will be responsible for designing and optimizing compiler components for AI models across various hardware platforms. The ideal candidate will possess deep knowledge of LLVM, MLIR and substantial experience with C/C++ programming.

The position offers a chance to work on cutting-edge technology and contribute to significant open-source projects.

Qualifications

• 7+ years experience in compiler engineering or related fields.
• Deep knowledge of LLVM and MLIR internals.
• Proven expertise in C/C++ programming.

Responsibilities

• Design, develop, and maintain critical compiler components.
• Optimize AI models for deployment across hardware platforms.
• Enhance PyTorch Inductor, Triton, LLVM, and MLIR toolchains.

Skills

Compiler engineering

C/C++ programming

Performance optimization

LLVM and MLIR internals

PyTorch

Triton

Debugging

Education

Bachelor’s or Master’s in Computer Science or related field

Job description

Role Description

We are looking for an experienced, highly skilled and motivated Senior Compiler Engineer to join our compiler team. In this role, you will be responsible for designing, developing, and maintaining critical components of our AI-driven compilation stack. You will work on PyTorch, Triton, LLVM, and MLIR to build robust, scalable, and high-performance solutions for diverse hardware backends. This is a hands-on technical position where you will solve complex problems, optimize for performance, and contribute to the next generation of our technology.

What You’ll Do

• Design, Implement, and Optimize Compiler Components: Architect and develop critical compiler modules to efficiently translate and optimize AI models for deployment across a variety of hardware platforms, including CPUs, GPUs, and emerging custom accelerators
• Enhance and unify PyTorch Inductor, Triton, LLVM, and MLIR toolchains to support cutting-edge architectures, facilitate seamless interoperability, and enable rapid experimentation with new compiler features
• Create and maintain custom IRs, code generation passes, and optimization strategies tailored for AI workloads, focusing on both general and domain-specific improvements
• Profile and tune computational kernels—such as linear algebra operations, matrix multiplications, and elementwise computations—to achieve optimal performance, scalability, and resource efficiency on diverse hardware
• Open-Source Engagement: Actively contribute to the LLVM, MLIR, Triton, and PyTorch open-source projects, sharing improvements, collaborating with the developer community, and driving the evolution of the AI compiler ecosystem

What We’re Looking For

• Bachelor’s or Master’s in Computer Science, Computer Engineering, or a related field from a recognized university
• 7+ years experience in compiler engineering or closely related fields
• Deep knowledge of LLVM and MLIR internals, including IR transformations, code generation, and backend optimization techniques
• Experience with PyTorch (Inductor/Dynamo) and Triton, including their compiler subsystems
• Proven experience and expertise in C/C++ programming
• Demonstrated expertise in performance optimization, including vectorization, parallelization, and hardware-specific tuning
• Advanced debugging, analytical, and system-level thinking skills
• Excellent communication skills with a strong track record of cross-functional collaboration

Ways to Stand Out from the Crowd

• Strong understanding of AI models — both training and inference pipelines
• Experience developing compiler support for custom hardware accelerators, including ASICs, FPGAs, or novel AI chips
• Active contributions to open-source compiler frameworks, demonstrating leadership and community involvement
• Familiarity with distributed training strategies, graph compilers, and advanced memory models