About

About Shubham

Hi, I’m Shubham. I’m a Senior Machine Learning Engineer at Meta, working on Instagram ads ranking. These days I’m particularly interested in multimodal LLMs and AI agents.

Previously, I spent two years at Zefr building multimodal search and zero-shot classification systems, and before that worked across computer vision roles at Triangulate Labs, UIUC, Reliance Jio, and Pitney Bowes — spanning skin cancer detection, hurricane damage estimation from satellite imagery, and building footprint extraction at scale.

I firmly believe that Specialization is for insects, which is why I like having multiple interests — technology, startups, education, philosophy, and football.

Education

University of Illinois at Urbana-Champaign — M.S. in Statistics (GPA 4.0)

• Coursework: Statistical ML, NLP for Deep Learning, Probability & Statistics, Computer Vision
• Research: Hurricane property damage estimation using satellite images + insurance data

University of Delhi — B.Tech (I.T. and Mathematical Innovations)

• Coursework: Computational Linguistics, Linear Algebra, Philosophical Implications of Technology, Micro-economics, Graph Theory

Skills

Machine Learning, Computer Vision, PyTorch, Python, OpenCV, AWS, Azure, Docker, C++, Django, Git, MongoDB, MySQL, Rust, TensorFlow, Java, Recommender Systems, Data Visualization, Image Processing, Text Classification