Vancouver, British Columbia, Canada
"Looking for CS internships | Open to Hackathon Team-ups" Hello! I’m Karan Anand, a driven Computer Science & Math student at the University of British Columbia with a passion for blending technology with practical solutions. My journey in tech has led me to exciting opportunities at UBC and Seaspan, where I honed my skills in software development, 3D visualization and VR/AR/MR. I’m currently expanding my expertise into the realms of AI and machine learning, with hands-on projects like developing AI-powered games and exploring web 3.0 technologies. Whether it’s through my academic projects or my professional roles, I aim to build systems that are not only innovative but also intuitive and user-friendly. Outside of coding, I enjoy playing badminton and sharing my journey and insights with peers through platforms like YouTube, hoping to guide and inspire future innovators. I’m always eager to connect with like-minded professionals and explore opportunities in software development, AI research, and tech innovation. Let’s connect and make a difference together!
I contribute to stdlib, an open-source standard library for JavaScript and Node.js focused on scientific and numerical computing. I began contributing in late 2024 by adding new statistical distributions. Over time, I took on more responsibility, reviewing pull requests, refactoring parts of the codebase, and helping maintain project quality. In May 2025, I was nominated as a Core Contributor after contributing ~200 PRs across JavaScript and C. I’m currently continuing my work through Google Summer of Code 2025, developing generalized universal functions (ufuncs) to expand support for core special mathematical functions in scientific computing.
🎓 YouTube Educator | Academic Assistance | Educational Content Creator 📚
Working under Prof. Christoph Ortner on the development of machine learning surrogates for particle models, including interatomic potentials and coarse-grained molecular dynamics. The research focuses on exploring various tensor formats within the (M)ACE architecture family to evaluate their expressivity, computational efficiency, and learning behavior in geometric deep learning. The goal is to ground design choices in mathematical and physical principles rather than convenience. Leveraging Python (JAX, PyTorch) and Julia to run experiments and develop scalable, interpretable models.