Rohil Shah

Hi, I’m Rohil. 👋

I am an incoming Master’s student at Stanford University in Computer Science. I am supported by the National Science Foundation Graduate Research Fellowship and have previously interned at Google, ChipAgents, and NetApp. I am currently interning at Roblox, working on search and recommendation systems.

I completed my undergraduate degree in Computer Science from the University of California, Santa Barbara within the College of Creative Studies. I was a student researcher in the UCSB Neural Information Retrieval Lab, advised by Prof. Tao Yang.

My research focused on the intersection of Systems and Machine Learning. I am interested in designing the data infrastructure and retrieval logic required to support reliable, large-scale Agentic AI systems.

At UCSB, I investigated efficient hybrid retrieval architectures: combining dense semantic embeddings with learned sparse representations to improve search quality without sacrificing efficiency. My daily work involved open-source vector search libraries like FAISS and experimenting with dynamic pruning strategies on the MS MARCO benchmark. I am broadly interested in how we can co-design retrieval algorithms with low-level systems infrastructure to handle the latency constraints of modern agents.

Previously, I interned at NetApp, where I built agents and RAG pipelines for root cause analysis on system logs. Before that, I worked at Inogen on medical device simulation, app development, and firmware testing. Beyond my technical work, I founded the UCSB ACM Student Chapter. Watching this community take root and learning from the incredible students, faculty, and staff at UCSB has been a highlight of my time here.

If you are interested in discussing efficient retrieval, AI systems, or distributed infrastructure, feel free to reach out!