My research lies at the intersection of data, language models, and AI agents. I focus on making data more accessible and useful to AI systems, particularly by developing methods to identify data that is unsuitable for language models and agents, understand why it fails, and automatically improve it.
I also develop more precise and reliable methods for evaluating AI systems, as well as multimodal approaches that combine speech and text models.

Previously, I was a postdoctoral researcher in the Natural Language Processing Lab at McGill University and Mila. I received my Ph.D. from the Natural Language Processing Lab at Bar-Ilan University. I have also conducted research at Amazon Research, IBM Research, and Intel’s AI Solutions Group.

I am hiring! I am currently looking for motivated M.Sc. and Ph.D. students who are excited about tackling challenging research problems in these areas. If you are interested in developing new methods for building and evaluating AI systems, please get in touch.

Misc

An interview about my research at Kan Tarbut radio channel (Hebrew)