Generative Glide: AI-driven ultra-large virtual screening for real-world drug discovery
Date & Time - June 4th, 2026
- 8:00 AM PDT | 11:00 AM EDT | 4:00 PM BST | 5:00 PM CEST
Location - Virtual
Accessing ultra-large virtual libraries is no longer the primary hurdle in small molecule discovery – extracting meaningful, purchasable hits from them is. Between heavy compute constraints, complex IT infrastructure requirements, and the diminishing returns of traditional docking, brute-forcing billions of compounds simply isn’t practical for most teams.
Generative Glide applies generative AI to overcome these challenges through goal-directed hit finding. Rather than exhaustively docking every molecule, the model intelligently navigates ultra-large chemical space to generate highly diverse, commercially available compounds that meet your specific project criteria. This targeted approach reliably identifies high-quality hits across distinct chemotypes, giving your project diverse starting points.
Crucially, this efficiency eliminates the need for massive compute clusters. You can now execute a full-scale, goal-directed campaign on a single modern workstation in just 2 days.
Join us as we go beyond slides and run a demo of the workflow, showing how Generative Glide performs in practice from setup through results. You’ll see what’s different, where it adds value, and where it fits (or doesn’t) in a real discovery setting.
Key Highlights
-
Introduction to Generative Glide: Our scientists will talk about how you can screen ultra-large libraries (up to 50B compounds) in under 2 days with Generative Glide
-
How does it compare: We will show benchmarking data to demonstrate how Generative Glide achieves improved top-hit quality and greater pose diversity as compared to AL-Glide
-
See it in action: Demo of Generative Glide – from setup through analysis of results
Who Should Attend
-
Computational chemists and molecular modelers
-
Scientists working in small molecule hit discovery or lead optimization
-
R&D teams exploring ultra-large virtual screening strategies
-
Drug discovery leaders seeking scalable, efficient screening solutions
Our Speakers
Kun Yao
Research Leader, Hit Discovery, Schrödinger
Dr. Kun Yao, product manager for machine learning-boosted physical simulation drug discovery workflows, joined Schrödinger after completing his graduate studies. He is responsible for overseeing the development of advanced computational platforms at Schrödinger, including Generative Glide and Active Learning workflows. Previously, Dr. Yao gained valuable applied biotech experience during a tenure at Scorpion Therapeutics. He obtained his PhD in computational chemistry from University of Notre Dame under the supervision of Prof. John Parkhill, where his research focused on combining machine learning with physical simulations, such as machine learning force field.
Steven Jerome
Global Portfolio Lead for Small Molecule Drug Discovery, Schrödinger
Dr. Steven Jerome is the Executive Director of Hit Discovery and Head of EU+ Application Science at Schrödinger. Currently, he also serves as the Global Portfolio Lead for Small Molecule Drug Discovery, where he oversees an expansive commercial portfolio of state-of-the-art computational tools, including advanced products like FEP+ and Glide. Steven initially joined Schrödinger as a scientific developer working on the Glide team before moving into product management. He holds a PhD in Chemistry from Columbia University, completed under Richard Friesner, where he developed tools for molecular docking and protein structure refinement.