Tugsuu M.

Close

Profile:

Tugsuu is a highly skilled software engineer with a deep interest in machine learning, infrastructure, and compilers. With a strong academic foundation from MIT, Tugsuu has contributed to various high-impact projects in both industry and academia. His work spans from optimizing parallel graph algorithms to enhancing the reliability and scalability of complex software systems. At Meta, Tugsuu applies his expertise to develop and optimize PyTorch compilers, ensuring high performance and efficiency. He is a dedicated mentor with a passion for advancing the fields of software engineering and AI.
Machine Learning
Software Engineering
Compilers
Infrastructure

Academic Background:

Master of Engineering in Artificial Intelligence, Massachusetts Institute of Technology (MIT)

  • Thesis focused on optimizing parallel graph algorithms by extending the Graphit DSL.
  • Graduated with extensive experience in machine learning and software engineering.

‍

Bachelor of Science in Computer Science and Engineering, MIT

  • Completed various projects in software development and AI.
  • Active participant in extracurricular activities, including leadership roles in student organizations.

‍

Professional Experience:

Software Engineer, Meta

  • Developing and optimizing PyTorch compilers, focusing on improving performance and scalability.
  • Contributing to infrastructure projects that support large-scale machine learning models.

‍

Graduate Research Assistant, MIT CSAIL

  • Worked on projects aimed at optimizing parallel graph algorithms, contributing to significant advancements in the field.
  • Collaborated with leading researchers in the COMMIT Compiler Group.

‍

Software Engineer Intern, Samara

  • Designed and deployed a custom zone-aware AWS load balancer to decrease compute costs.
  • Focused on software reliability and infrastructure optimization.

‍