Stanford University
Darve Lab
Reliable AI for Science and Engineering
The research group of Eric Darve, Director of the Institute for Computational and Mathematical Engineering (ICME),
Professor of Mechanical Engineering and, by courtesy, of Computer Science at Stanford University.
People
Research Team
Research
Reliable AI for Science and Engineering
The Darve Lab develops algorithms and AI systems that make scientific computation faster, experimental data easier to use, and results easier to understand and verify. Our research connects numerical methods, machine learning, and scientific applications.
Fast Algorithms for Data Processing at the Edge
We develop algorithms that compress and analyze data close to where it is generated, including directly at scientific detectors.
Computational ImagingNumerical Methods & HPCAI Agents for Science
We build AI workflows that help researchers translate scientific questions into simulations, diagnose failures, and assess whether the results are physically and numerically sound.
Agentic AIAI for ScienceLanguage-Model Interpretability and Alignment
We study how large language models represent knowledge and use it during computation.
AI InterpretabilityAI SafetyScientific Machine Learning for Faster Simulation and Physical Insight
We combine physical models with machine learning to reduce computational cost and understand complex systems.
AI for ScienceSciML
News
Lab updates
Eric Darve’s courtesy appointment in Computer Science approved
Prof. Eric Darve’s appointment as Professor of Computer Science, by courtesy, at Stanford University was approved on August 5, 2026.
Darve Lab receives DOE Genesis Mission award for ORACLE
Led by Prof. Eric Darve, ORACLE will develop AI agentic workflows for analyzing large experimental datasets in particle physics, with a focus on the Deep Underground Neutrino Experiment (DUNE). The project brings together Stanford and SLAC researchers through the Department of Energy’s Genesis Mission.
Eric Darve joins Genesis Mission project on subsurface energy
Prof. Eric Darve is co-PI on a DOE Genesis Mission project led by Prof. Hamdi Tchelepi. The team will develop specialized AI agents to simulate underground rock fractures, check results against physical laws, and quantify uncertainty, supporting geothermal and other subsurface energy applications.
Celebrating Stanford and ICME’s Class of 2026
Congratulations to the Stanford and ICME graduates of 2026! We celebrate their accomplishments and wish them success in their next steps in research, industry, and beyond.
ICME Research Symposium explores AI for Science and Engineering
The 2026 ICME Research Symposium brought together faculty, students, and industry partners for talks, discussions, and student posters on AI for Science and Engineering. Prof. Eric Darve delivered opening and closing remarks and participated in a fireside discussion.
ICME Career Forum connects students with industry and national labs
The 2025 ICME Career Forum brought Stanford students and faculty together with representatives from industry and national laboratories for a day of networking, interviews, and career conversations at the Huang Engineering Center.
Darve Lab receives funding for LLM interpretability and alignment
The Darve Lab received an award from Open Philanthropy to support research on large language model interpretability and alignment. The organization is now known as Coefficient Giving.