Open postdoc position: Geometric Approaches to Safe and Robust AI

[Justin Solomon]

I'm delighted to share an open postdoc position hosted jointly between the Geometric Data Processing (GDP) group in the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) and the MIT-IBM Computing Research Lab, mentored by me, Ankit Gupta, Leshem Choshen, and friends; the start date can be as early as fall 2026.

The broad theme of our collaboration is "Geometric Approaches to Safe and Robust AI," with topic-wise flexibility. Example areas of interest include data generation, gradient-free optimization, continual learning, geometry of fine-tuned models, geometry of learned embeddings, and diffusion-based sampling. We are particularly excited for candidates who reach out with research directions of their own in mind that they can link to our ongoing work.

Please share with your students and colleagues! And if you are interested to apply, please reach out ASAP to me (jsol...@mit.edu) with your CV and a *short* description of the kind of research you might do in our lab. Please keep things short and refrain from sending AI-generated "slop" as part of your inquiry.

- Justin Solomon
Associate Professor of Electrical Engineering and Computer Science
MIT Computer Science and Artificial Intelligence Laboratory