If you might be interested in joining us, please take a look at our group guide for information about our values and my attitude toward research and mentorship.

Please note that due to the high volume of email, I may not be able to respond to every message I receive. If you don’t see a response after 1-2 weeks, please feel free to follow-up with a second message.

Interested postdocs should inquire directly at ccoley@mit.edu with a CV and a brief statement of your research interests and background. Please also mention what your professional goals are, and what you would hope to learn and accomplish, specifically, while working in the group. Even if we do not have any active postings, I will keep your materials on file and reach out if specific openings arise at a future date.

Graduate students should apply directly to the MIT Department of Chemical Engineering or the Department of Electrical Engineering and Computer Science (note: fee waivers are available through the Dean of Engineering) or the Department of Chemistry. We also have had students join from the Computational Science and Engineering program and the Computational and Systems Biology program. Admissions is handled department-by-department or program-by-program, and as an individual I do not have full discretion to admit students without committee support. As a general rule, I plan to recruit at least two PhD students annually; the exact projects will vary and should reflect our mutual interests and funding.

Undergraduate students looking for research opportunities for pay or credit are encouraged to directly reach out to graduate students and postdocs in the group whose interests match their own. You can also email ccoley@mit.edu with your interests and any relevant coursework or research experience. However, please keep in mind that I may receive dozens of such inquiries each day, and so I am in general unable to reply to every message.

At the moment, we are particularly interested in candidates who can help lead our work in synthesizability-constrained molecular generation, which is a molecular design task posed as the generation of a directed acyclic graph. This work requires a good understanding of deep learning for tree/graph generation and would benefit from a familiarity with molecular machine learning.