Senior ML Compiler Engineer

About Fractile

Fractile is pioneering a groundbreaking computing approach aimed at running the world's largest language models at speeds 100 times faster than current systems. Our rapidly expanding team operates at the forefront of AI advancements across both hardware and software domains.

Role Overview

We are recruiting Senior Machine Learning Compiler Engineers to contribute to the development of the compiler stack tailored for our state-of-the-art AI accelerators. Candidates may work from our Bristol or London offices, based on preference.

Why Join Us?

Fractile is not simply another AI startup connecting existing frameworks. We are creating a comprehensive full-stack execution engine, including hardware, compiler, and runtime components, to dramatically enhance the performance of the largest language models. This ambitious endeavor involves reimagining the entire system from fundamental principles.

For compiler engineers, this is a unique opportunity: with architecture still being finalized, you will influence the design of the IR, lowering strategies, optimization passes, and execution models.

Key Responsibilities

• Construct and enhance the machine learning compiler stack for Fractile’s cutting-edge AI acceleration hardware
• Optimize performance of current large language models on Fractile’s accelerator platform
• Collaborate closely with hardware, driver runtime, and machine learning teams using a hardware-software co-design approach

Required Qualifications and Skills

• Demonstrated expertise in developing machine learning compilers utilizing industry-standard frameworks like MLIR or TVM
• Current understanding of machine learning methodologies and technological advances
• Strong grasp of computer architecture and techniques for performance optimization
• Proficiency in C and C++ programming languages along with experience using standard development tools and environments
• A proactive, inventive attitude with capability to take initiative and deliver in a dynamic setting
• Degree in Computer Science, Electronic Engineering, Mathematics, Physics, or a related field plus over three years of professional industry experience

Additional Preferred Experience

• Background in startup or small, agile team environments