I am interested in the application of imitation learning, reinforcement learning, and deep learning to algorithmic domains with strong heuristic aspects, such as optimization and combinatorial search. I tend to look at applications within machine learning, natural language processing, and computer vision. I am also interested in the software engineering side of machine learning, where I like to think about modularity, reusability, expressivity, workflow, and tooling. I think that, as a community, we should pay a lot more attention to interfaces, standardization, and design patterns.
I am currently contracting part-time for Microsoft to continue working on the research direction that we have pursued during my last internship at Microsoft Research. From January 2018 to May 2018, I was a research scientist intern at Petuum working on an architecture search framework (to be open sourced soon!). I have spent two summers at Microsoft Research: one in 2018, working with Ben Zorn and Alex Polozov on applying deep learning to spreadsheets; one in 2016, working with Asela Gunawardana and Vincent Etter on machine learning for user interaction. Before coming to CMU, I spent the summer of 2015 as a research intern at Xerox Research Centre Europe working with Ariadna Quattoni and Xavier Carreras on sketching algorithms for spectral learning. I spent the year of 2014 as a research scholar working with André Martins on machine learning and natural language processing. I spent three months as a visiting scholar at CMU working with Noah Smith on multiword expression tagging. Prior to that, I got a MSc and a BSc in Electrical and Computer Engineering from Instituto Superior Técnico. For my MSc thesis, I worked with Pedro Aguiar on developing invariants for shape representation using symmetric polynomials and the bispectrum. It has been a pleasure to work with so many great people!
I am always happy to discuss research. Send me an email or stop by my office if you want to chat.