AI-Powered Biological Software with Jakob Uszkoreit, CEO of Inceptive
Aug 24, 2023
35:23
auto_awesome Snipd AI
Jakob Uszkoreit, co-founder of Inceptive and former Google Translate and Assistant teams lead, discusses the application of deep learning in biological design, optimizing vaccine production, and more efficient drug discovery. Topics covered include the use of attention in deep learning, exploring potential architectures, challenges in AI systems, deep learning in biological research, language pre-training, and the iterative process of data generation in AI-powered biological software.
Read more
AI Summary
Highlights
AI Chapters
Episode notes
auto_awesome
Podcast summary created with Snipd AI
Quick takeaways
Inceptive is using deep learning to design RNA and mRNA molecules for more efficient drug production and distribution in the field of medicine.
The application of deep learning to biology enables the design of complex RNA-based medicines with personalized functions and scalability advantages for future medical interventions.
Deep dives
Inceptive's Focus on Biological Software
Inceptive, a company founded by Jakub Uskarit, aims to design better RNA and mRNA molecules for various medicines. The goal is to use deep learning models to translate biological programs into RNA molecule descriptions that can perform specific functions in the body. This approach allows for the creation of complex and scalable medicines that can be manufactured and distributed more efficiently than traditional protein-based biologics. Inceptive believes that deep learning, with its ability to work with black box systems, could revolutionize the field of biology and lead to significant advancements in medicine.
The Potential of RNA and mRNA
RNA and mRNA molecules have a wide range of potential applications, especially in the field of infectious disease vaccines. With deep learning, it is possible to design RNA molecules that can carry out specific tasks, such as protein production. The ability to program and manipulate RNA opens up possibilities for personalized medicine and the development of complex medicines with functionalities like conditional statements and recursion. RNA-based medicines also offer advantages in terms of manufacturing scalability and distribution, making them a promising modality for future medical interventions.
The Role of Deep Learning in Biology
The application of deep learning to biology is an emerging field that combines the principles of deep learning with biological systems. This interdisciplinary approach allows for the exploration of new research areas and the development of innovative solutions. Inceptive views itself as part of a new discipline that draws from both deep learning and biology. By integrating deep learning models and biological experimentation, the company seeks to optimize the design of experiments and assays, creating a symbiotic relationship between wet lab experiments and in silico modeling.
The Collaborative Nature of Inceptive's Approach
Inceptive promotes a collaborative and interdisciplinary environment where experts from various fields, such as deep learning, robotics, and biology, work together. This interdisciplinary collaboration leads to unique problem-solving approaches and the discovery of solutions that traditional disciplinary boundaries may have restricted. By fostering an environment where different languages and perspectives converge, Inceptive aims to harness the magic of cross-disciplinary collaboration and achieve groundbreaking advancements in the field of biology.
"Biological Software" is the future of medicine. Jakob Uszkoreit, CEO and Co-founder of Inceptive, joins Sarah Guo and Elad Gil this week on No Priors, to discuss how deep learning is expanding the horizons of RNA and mRNA therapeutics.
Jakob co-authored the revolutionary paper Attention is All You Need while at Google, and led early Google Translate and Google Assistant teams. Now at Inceptive, he's applying these same architectures and ideas to biological design, optimizing vaccine production, and magnitude-more efficient drug discovery. We also discuss Jakob's perspective on promising research directions, and his point of view that model architectures will actually get simpler from here, and be driven by hardware.