I'm a research scientist whose work lies at the intersection of neuroscience, cognitive science, and AI. I am primarily interested in understanding

(1) how minds, brains, and machines represent information about objects, concepts, and categories;

(2) how the form of these representations (i.e. representational geometry and topology) help achieve various cognitive functions;

(3) how systems with different learning mechanisms, like biological and artificial neural networks, learn, generalize, and develop representations with similar properties; and 

(4) how advances in mechanistic interpretability of machine learning models can enhance building safe, transparent, and reliable AI systems.

I am currently a postdoc in the Bytes of Minds Lab in the Windreich Department for AI & Human Health at Mount Sinai and received my PhD in Computational Cognitive Neuroscience from Rutgers University working in the Mind Brain Analysis group and the Brain Imaging Center. Previously, I was also a joint-postdoc in the Concepts and Actions Lab at Carnegie Mellon University and in the ProAction Lab at the University of Coimbra. Before that, I spent multiple years researching how children acquire and apply causal reasoning and causal inference in the Computational Cognitive Development Lab and at the Central European University's Cognitive Science Department. Once upon a time, I also used to be a linguist and received a MA in linguistics from the Utrecht University. 

Additionally, I am a big advocate for responsible AI and AI transparency, which can only happen if we truly understand what the system's internal mechanisms are.

Nov 2026 I gave the introductory talk for the new public series I'm leading, Show Me the Data: The Benchmark Problem, on the AI & Equality and Human Rights community platform of Women at the Table. It looks at what AI benchmarks actually measure, and what they miss.

Oct 2026 Excited to be chosen as one of the teaching assistants at the Simons Computational Neuroscience Imbizo 2027!

Sep 2026 Our paper Directional confusions reveal divergent inductive biases through rate-distortion geometry in human and machine vision has been accepted at NeurIPS 2026. See you in Sydney in December!

Aug 2026 New preprint out on bioRxiv: Geometric reshaping of task-relevant representations in the primary visual cortex supports perceptual decisions.

Jul 2026 Invited talk on rate-distortion signatures of generalization in brains and machines at the CNS*2026 Workshop on Methods of Information Theory in Halifax, Canada.

Jul 2026 Co-organized the London Geometry and Machine Learning Summer School (LOGML), which brings together early-career researchers working at the intersection of geometry, topology and machine learning.

May 2026 Co-organized the AI & Equality: Festival of Ideas conference and hosted three panels.

Mar 2026 Presented two posters at COSYNE 2026 in Lisbon: one on rate-distortion signatures of generalization in humans and deep networks, and one on how learning reshapes geometry in mouse visual cortex.