I am a computer science researcher. In 2027, I will join Carnegie Mellon University as an Assistant Professor in CSD (and HCII by courtesy), affiliated with the CMU Database Group. You can find my group's blog: Full Stack Data Lab. See my Google Scholar for my papers.
My PhD is in EECS from UC Berkeley, where I worked with Aditya Parameswaran and built the DocETL stack for LLM-powered data processing (3.7k+ GitHub stars, used by public defenders, climate scientists, and more). We won several paper awards for this line of research (Best of SIGMOD 2026, Best Paper at CHI 2026, Best Paper Honorable Mention at UIST 2025). In a past life, I was an ML engineer, and my undergrad is in CS from Stanford.
I also teach a popular course on AI evals with my friend Hamel Husain, and we wrote a book together. We are so grateful for how much students love the course. Many of our materials are freely available: an evals FAQ, some of our videos, and memes (one for every scenario, I promise).
Current mentees
- Arnav Dhariya (undergrad @ UC Irvine)
- Lindsey Wei (undergrad → PhD student @ Berkeley; CRA Undergraduate Award Honorable Mention)
Past mentees
- Parth Asawa (undergrad → PhD student @ UC Berkeley; CRA Undergraduate Award Honorable Mention)
- Ruiqi Chen (MS → PhD student @ University of Michigan CSE)
- Andrew Cheng (undergrad → MS @ UC Berkeley)
- Ankush Garg (MS → Senior Data Scientist @ Clarkson Consulting)
- Rachel Lin (undergrad, MS → Software Engineer @ Opto)
- Aditi Mahajan (undergrad → Google)
- Nikhil & Vinay Rao (high school → undergrads @ UC Berkeley EECS)
- Quentin Romero Lauro (undergrad → CEO @ Inspector, YC 2025; CRA Undergraduate Award Winner)
- Sasha Singh (undergrad → Software Engineer @ Stripe)
- Reya Vir (undergrad → PhD student @ Columbia; NSF GRFP recipient)
- Yujie Wang (undergrad → Google)
Recent news
- We released Quail, our new query-aware inference engine for AI-SQL workloads, and it processes over one billion tokens per minute on a single H100! Read more on the Full Stack Data Lab blog and the Modal blog.
- Our paper on Data Agent Bench (DAB), a new benchmark for data agents, will appear at EMNLP in October! DAB has already received over 50 submissions.
- We wrote What Do Users Actually Do with LLM-Powered Data Systems? for the June 2026 issue of the IEEE Data Engineering Bulletin.
- Two papers, Multi-Objective Agentic Rewrites for Unstructured Data Processing and Featurized-Decomposition Join, appeared at VLDB 2026.
- Two papers, Cut Costs, Not Accuracy and Task Cascades, appeared at SIGMOD 2026! Cut Costs, Not Accuracy received a Best of SIGMOD 2026 award.
- RAG Without the Lag won a Best Paper Award at CHI 2026!
Academic service
Reviewer/Program Committee Member: CIDR (2027–), VLDB (2027–), UIST (2024–), CHI (2024–), NeurIPS (2021, 2022)
Organizer: DEEM Workshop at SIGMOD (2023–2025)