Ramin Hasani is an artificial intelligence researcher and entrepreneur. He is Co-founder and Chief Executive Officer of Liquid AI, a company focused on developing foundation models and adaptive AI systems. Ramin Hasani has been affiliated with the Massachusetts Institute of Technology, specifically the Computer Science and Artificial Intelligence Laboratory (CSAIL).
He is known for his work on Liquid Time-Constant (LTC) networks. LTC networks are a class of recurrent neural network models designed for continuous-time dynamics. These models were introduced in peer-reviewed research and are sometimes called “liquid neural networks” in media coverage. His research has focused on neural network architectures that adapt to time-varying inputs and operate in dynamic environments.
Research Contributions
Liquid Time-Constant networks were developed to address limitations in traditional recurrent neural networks. The architecture uses differential equations that let neuron states evolve continuously over time.
Published research on LTC networks has demonstrated their application in time-series prediction and control tasks. Academic studies have included autonomous-driving simulations and other dynamic-system environments. The work has been cited in the academic literature on neural network efficiency, interpretability, and adaptive modeling.
Founding of Liquid AI
Ramin Hasani co-founded Liquid AI to develop and commercialize adaptive AI systems based on continuous-time neural network research.
Liquid AI focuses on building foundation models and AI systems for deployment in real-world environments. The company’s work centers on efficiency, adaptability, and practical implementation of advanced neural architectures, and serves as Chief Executive Officer, and is responsible for the company’s strategic direction and operations.
Technical Focus
Liquid AI’s work includes developing neural network architectures with improved parameter efficiency compared to some traditional deep learning models, depending on the application. The company emphasizes models that adapt to changing input conditions and operate in environments where computational resources may be constrained. The research behind these systems builds on continuous-time modeling and adaptive neural dynamics.
Publications and Recognition
Ramin Hasani has co-authored peer-reviewed papers on Liquid Time-Constant networks and related neural architectures. The research introducing LTC networks has been covered by academic institutions and technology media outlets for its proposed advantages in adaptability and efficiency. Its work is associated with ongoing research in neural network design and next-generation AI systems.
Current Role and Focus
As CEO of Liquid AI, Ramin Hasani leads the company’s efforts to develop and deploy adaptive AI foundation models. His responsibilities include overseeing research translation into commercial products, guiding technical development, and managing company operations.
Liquid AI continues to work on scalable AI systems for integration into real-world applications.
Ramin Hasani is an AI researcher at MIT CSAIL and the co-founder of Liquid AI. He is recognized for his work on Liquid Time-Constant neural networks, a continuous-time neural architecture for adaptive modeling. liquid AI, he is advancing research-based neural network architectures toward commercial AI systems that prioritize adaptability and efficiency.
