About me

I recently completed my Ph.D. at the Quebec Artificial Intelligence Institute (Mila) under the guidance of Professor Laurent Charlin. My academic journey has been marked by collaborations with globally renowned institutions, including DeepMind, where I worked within the Continual Learning team led by Marc’Aurelio Ranzato, Amazon in Alex Smola’s team, and now at ServiveNow as a visiting researcher. I also had the privilege of contributing to ElementAI before its integration with ServiceNow.

At the heart of my research is the development of algorithms proficient in accumulating and transferring knowledge or skills to enhance generalization across varied tasks. My passion for data and computational efficiency has directed my studies into continual, transfer, and meta-learning, with a particular emphasis on applications spanning language, vision, and reinforcement learning.

Currently, at ServiceNow, I’m applying these principles by working with pre-trained Large Language Models (LLMs) to create computer task-solving agents. These agents, leveraging the vast knowledge and adaptability of LLMs, aim to navigate and master various computer tasks more efficiently. This work represents an exciting step towards enhancing the capabilities of automated solutions in different sectors.

Also, check out our software to unify continual-learning research, Sequoia, as well as my continual-learning wiki.