About me

I am on the job market this cycle for faculty or industry research positions starting in Fall 2024. My research agenda centers on embodied lifelong learning. I cover my thoughts on how we can tackle this massive challenge in this blog post. Here is my research statement.

I am a Postdoctoral Fellow at the Learning & Intelligent Systems Group at MIT CSAIL, where I work with Prof. Tomás Lozano-Pérez and Prof. Leslie Pack Kaelbling. I obtained my Ph.D. at the Lifelong Machine Learning Group at the University of Pennsylvania, where I was advised by Prof. Eric Eaton. I previously obtained a Master’s degree in Robotics from the GRASP Lab at Penn, and got my Bachelor’s degree in Electronics Engineering from Universidad Simón Bolívar in Venezuela.

During my Ph.D., I was fortunate to work as a research intern in 2021 at FAIR NYC with Arthur Szlam and Ludovic Denoyer on mixture-of-experts models, in 2020 at MSR Montreal with Harm van Seijen on lifelong learning of compositional reinforcement learning problems, and in 2019 at Facebook AI with Alborz Geramifard and Mohammad Ghavamzadeh on multidomain reinforcement learning for dialog. I previously spent a year at Politecnico di Milano in Italy, where I took graduate-level courses in AI and Robotics, and spent five months at the Advanced Systems Technology group in STMicroelectronics.

Research Interests

I’m interested in the creation of versatile, intelligent, embodied agents that learn to better interact with the world by accumulating knowledge over their lifetimes. I focus on the question of how these agents can leverage various forms of compositional and modular structures to transform the complex problem of modeling a lifelong data stream into simpler problems that can be more easily solved and whose solutions can be adapted, recombined, and reused in the future. My work mainly applies these methods to robotics, and also engages with related fields like computer vision and natural language.