ML Compiler Engineer

Looking for an ML Compiler Engineer!

This exciting start-up focuses on harnessing the potential of AI for the betterment of humanity. In pursuit of this goal, they are crafting a comprehensive solution encompassing both software and hardware, engineered to deliver vastly superior performance and efficiency compared to traditional methods. The design emphasizes scalability, ensuring adaptability to evolving needs

Currently, they are in the process of creating the SPU (Spatial Processing Unit), an economical AI processor meticulously crafted from foundational principles to offer unparalleled performance per watt, empowering the advancement of next-generation AI workloads.

Skills:

• Strong proficiency in organization and time management

• Expertise in Software Architecture

• Proficient in Parallel Programming

• Exceptional proficiency in C/C++

• Extensive experience with CUDA

• Advanced Problem-Solving Skills

• Demonstrated proficiency in PyTorch with successful deployment of models

• Experience in performance and memory optimization

• Aptitude for benchmarking and optimization techniques

Requirements:

• A bachelor's degree in a STEM field (preferably EE or CS). Candidates with a master's degree or equivalent experience will be given preference.

• A deep understanding of algorithms and numerical analysis

• Familiarity with low-level software drivers and kernel operations

• 4+ years of experience in software development for products and/or enterprise

• Some knowledge of compiler design and development

• Experience working on large and complex applications, with proficiency in code manipulation and optimization

• Excellent problem-solving skills, with the ability to troubleshoot and resolve critical technical issues independently

• Proficiency in code versioning tools such as Git, Mercurial, or SVN

• Familiarity with continuous integration practices

• Knowledge of frameworks like TensorFlow, PyTorch, or ONNX is not mandatory but would be beneficial.