AI Engineers: The Next Generation — Stefania Druga, Google Gemini

Links for more info: https://stefania11.github.io/ Sign-up for early access here http://cognimatescopilot.com This talks explores all the ways in which we can push the boundaries of our learning and creativity by using a myriad of open-source multimodal agents that are proactive and nudge you to explore different topics, dimensions, approaches to your learning workflow while also pointing certain cognitive biases or misconceptions you might have. I will show some examples of mobile agents I created in Math learning using our math misconceptions benchmark, also examples in science using the Mozart misconceptions benchmark, an a series of reasoning and logical puzzles where a mobile proactive agent can support us to learn or sometimes un-learn ( for inspiration see all the experiments of Daniel Kahneman and Amos Tversky) Recorded live at the Agent Engineering Session Day from the AI Engineer Summit 2025 in New York. Learn more at https://ai.engineer and purchase tickets to our next event, the AI Engineer World's Fair, in SF June 3 - 5 here: https://ti.to/software-3/ai-engineer-worlds-fair-2025 About Stefania Hi! I am Stef. I am currently a research scientist in Google Gemini team working on real-world applications for multimodal language models. I was a principal researcher at the Center of Applied AI Research at the University of Chicago and graduated with a Ph.D. in Creative AI Literacies 🎓 at the University of Washington Information School. In the past I worked with Microsoft’s Human-AI eXperience Team, X Moonshot Factory, Fixie.ai. I also enjoy designing and building an AI coding plaftorm for kids which started as my master project at MIT (cognimates.me). When I am not coding & writing papers 👩🏽‍💻 I love trail running and rocket yoga.