In this talk, we embark on a journey to explore the unique properties of the latest Generative AI systems, highlighting how their intelligence differs from human intelligence in significant ways. We then walk through the practical applications of this AI across various domains and tasks, providing a framework for understanding its current capabilities and limitations. Lastly, we apply the framing to software development and talk through how we should be rethinking software development.

I am an AI researcher, engineer and community builder. I am currently at Meta and NYU, jumping between Engineering, Research and Leadership as I find convenient. I currently lead PyTorch and AI Infra-related things at Meta. These days I increasingly dabble with robots, at NYU. I specialize in machine learning platforms and generative models.

For a longer version, read more about me.