The Kulik Research Group, led by Professor Heather J. Kulik in the Departments of Chemical Engineering and Chemistry at MIT, is a prominent research team specializing in computational chemistry, machine learning, and materials discovery. The group develops advanced electronic structure methods and atomistic simulations to accelerate the design of catalysts, metal-organic frameworks (MOFs), and bio-inspired materials. With a strong emphasis on interdisciplinary collaboration and inclusivity, the Kulik Group has made significant contributions to fields such as catalysis and biochemistry. Of particular note is their pioneering work in agentic science, which leverages AI agents for autonomous scientific discovery. This report outlines the group's structure, research foci, and achievements, highlighting their role in advancing agentic paradigms.
The Kulik Research Group operates at the intersection of computational chemistry, chemical engineering, and materials science within MIT's Chemical Engineering Department, with affiliations to the Chemistry Department, Center for Computational Science and Engineering, and Computational Systems Biology program. Founded and led by Professor Heather J. Kulik, the group fosters an inclusive, collaborative environment that prioritizes member well-being and creative problem-solving. Kulik, who holds the Lammot du Pont Professorship, received her B.E. from Cooper Union (2004) and Ph.D. from MIT (2009), followed by postdoctoral positions at Lawrence Livermore and Stanford before joining MIT in 2013. She was tenured in 2021 and promoted to Full Professor in 2024.
The group comprises approximately 20 members, including postdocs (e.g., Beck Miller, Daniel Mukasa, Ethan Curtis), graduate students (e.g., Xiao Huang, Husain Adamji, Aaron Garrison), undergraduate researchers (e.g., Brian Ma, Natalie Kozlowski), and visitors (e.g., Giorgia Brosio, Hongliang Xin). An administrative assistant, Hannah Cross, supports operations. Collaborations extend to multidisciplinary centers like the Center for Enhanced Nanofluidic Transport (CENT) EFRC, SciDAC-5, and NSF Center for the Chemistry of Molecularly Optimized Networks (MONET). The group maintains an active presence on X under @KulikGroup, with over 3,780 followers, sharing updates on inorganic design, computational chemistry, machine learning, and enzyme catalysis.
Contact information includes: email at hjkulik@mit.edu, address at 77 Massachusetts Ave., Room 66-464, Cambridge, MA 02139.
The Kulik Group's research emphasizes multi-scale modeling, electronic structure calculations, and machine learning to discover new molecules, materials, and mechanisms. Core areas include:
Recent publications from 2025 include "The specificity and structure of DNA cross-linking by the gut bacterial genotoxin colibactin" (Science), "Active Site Dynamics in Molybdenum-Based Silica-Supported Olefin Metathesis Catalysts" (J. Am. Chem. Soc.), and "Excitonic Anisotropy in Single-Crystalline 2D Silver Phenylchalcogenides" (Adv. Opt. Mater.). The group's work addresses real-world challenges in sustainable materials and energy.
Agentic science, a key interest area, represents an emerging paradigm where AI agents semi-autonomously conduct research by reasoning, planning, and interacting with tools and environments. Professor Kulik co-authored a seminal commentary in Nature Machine Intelligence (September 2025) with Hongliang Xin (Virginia Tech) and John R. Kitchin (CMU), titled "Towards agentic science for advancing scientific discovery." The abstract highlights how AI transforms discovery through autonomous agents, discussing foundations, frontiers, limitations, and responsible integration.
Kulik envisions agentic science accelerating innovation by automating tedious tasks and enabling novel hypotheses, particularly in materials design and catalysis. This builds on the group's expertise in autonomous workflows and ML-accelerated discovery, such as bypassing trial-and-error in transition metal complexes and MOFs. Challenges include AI limitations like hallucinations and the need for human oversight, with calls for cross-institutional collaboration to ensure ethical deployment. The paradigm positions AI as a partner in self-driving labs, with potential impacts in environmental and health sciences.
The Kulik Group stands as a leader in computational materials science and AI-driven discovery at MIT, with agentic science exemplifying their forward-thinking approach to revolutionizing research methodologies. By integrating machine learning with autonomous systems, the group tackles complex challenges in catalysis and beyond, promoting responsible AI use. Future directions may expand self-driving workflows, enhancing scientific efficiency. For more information, visit the official website at http://hjkgrp.mit.edu/.