Director of MIT CSAIL, autonomous systems pioneer
Daniela Rus
Profile
Daniela Rus directs MIT CSAIL — the largest research lab at MIT and one of the most influential AI and CS labs in the world. She’s the first woman to run it, a MacArthur Fellow, and the 2025 IEEE Edison Medalist. For developers trying to understand where AI meets the physical world, her lab is where a huge amount of that future is being built.
Her own research sits at the intersection of robotics and machine learning. She’s a foundational figure in soft robotics — robots made from silicone, textiles, and other compliant materials instead of rigid metal. Her 2015 Nature paper with Michael T. Tolley, Design, Fabrication and Control of Soft Robots, is the field’s canonical reference. Her group has built swimming robotic fish, self-folding origami robots, and self-reconfiguring modular systems — hardware that looks nothing like the robots you grew up imagining.
On the software side, she’s the co-creator of liquid neural networks — a continuous-time architecture developed with Ramin Hasani, Mathias Lechner, and Alexander Amini. The headline result: a car can stay in its lane using 19 neurons instead of 100,000, and the network’s attention looks at the road horizon like a human driver does. That work spun out of MIT in late 2023 as Liquid AI, one of the more technically distinctive bets against the transformer monoculture.
What makes Rus worth paying attention to isn’t any single project — it’s that she sees robotics and AI as one problem. Under her leadership, CSAIL has become the place that takes “physical intelligence” seriously as the next phase of AI, and she argues it clearly to non-specialists. If you’re a developer wondering what happens when LLMs leave the chat box, her work is a direct line to the answer.
Books
The Heart and the Chip: Our Bright Future with Robots Rus's 2024 synthesis of where robotics is heading — accessible, opinionated, and grounded in decades of lab work.Key Articles & Papers
Design, Fabrication and Control of Soft Robots Liquid Time-constant Networks Robust flight navigation out of distribution with liquid neural networks Closed-form continuous-time neural networks "Liquid" machine-learning system adapts to changing conditionsVideos
Spotify Podcasts