Kuber Shahi

I’m a second-year Master’s student in Computer Science at UC San Diego, specializing in Artificial Intelligence. My current research focuses on LLM-based agents for complex tasks, especially improving multi-step reasoning and constraint handling. My broader research interests span Embodied AI, reinforcement learning, NLP, and their applications in healthcare and beyond.

Before grad school, I worked as a Data Scientist at Vayana Network, where I built scalable ETL pipelines, developed graph-based tools to uncover customer insights, and led NLP systems for deduplication of company records. I also built an internal reporting platform using Apache Superset to improve access to business metrics across teams.

I earned my Bachelor’s in Computer Science with a minor in Physics from Ashoka University. During my time there, I had the opportunity to explore research in privacy-preserving machine learning and applied cryptography with Professors Mahavir Jhawar and Debayan Gupta. I worked on topics such as secure neural network training, adversarial attacks on ML models, and privacy vulnerabilities in encrypted email systems.

Along the way, I’ve explored projects in abstractive text summarization, one-shot learning, and full-stack web development. I enjoy working on problems that require both creativity and technical depth, especially where real-world impact is involved.

Outside of work and research, I enjoy watching science fiction, mystery, and thriller movies. I’m a big Chelsea FC fan and follow football closely, as well as cricket and Formula 1 occasionally. I also like to swim and run during my free time.