Yunan Luo is an Assistant Professor in the School of Computational Science and Engineering at Georgia Institute of Technology. He is broadly interested in computational biology and machine learning, with a focus on developing AI methods to reveal core scientific insights into biology and medicine. His research has been recognized with several faculty awards, including the NSF CAREER Award and the NIH MIRA Award.

FutureHouse is a non-profit founded in 2023 to automate the intellectual tasks of science. We are automating each stage of scientific discovery – from hypothesis generation, data analysis, and literature search. We have announced our first major results on exceeding human level performance in summarizing and synthesizing literature, building a benchmark for biology tasks, and scientific agents in closed loop discovery. We have exceeded human level performance of experts in multiple tasks and recently released open source code and launched a platform to enable others to use these agents to accelerate protein design, literature research, disease-target interactions, and bioinformatics analysis. In this talk, I will review these results and give a general update on our progress, with specific results on automating synthesis of human knowledge, exceeding human level performance and frontier models with local models, and assessing the progress on frontier AI models for doing scientific tasks – like ether0, our chemistry reasoning model.

Andrew is cofounder and head of science at FutureHouse, where he has led multiple AI for science projects at FutureHouse including ChemCrow (first LLM agents in chemistry), ether0 (first reasoning model in a scientific domain), and paperqa (first superhuman literature agent). Andrew has been a professor and researcher in ML in chemistry, explainable AI, statistical mechanics, and chemical engineering, and has received numerous awards, including junior investigator awards from the NSF and NIH.

The type of problems to which AI is applicable has expanded at the same, drastic rate as its computational complexities. Machine-learned surrogates have progressed from thousand-parameter models with limited applicability to massive models that rival semi-empirical theory. Chemical design with AI has progressed from human-lead endeavors with single-task models to general reasoning AI engaging with the human world autonomously. NVIDIA and our partners are building tools that span from accelerating the core operations of surrogate models through toolkits redesigned for AI-powered simulation to the infrastructure which connects agentic AI to the physical world. This talk will introduce tools designed for different aspects of chemical research and how they are being used by our partners to drive science.

Logan Ward is an application engineer in the AI for Science team at NVIDIA. He helps teams in the public and private sectors build applications that weave together advanced AI and simulation techniques. Logan has a PhD in Materials Science and Engineering from Northwestern University and was a staff scientist at Argonne National Laboratory for six years before joining NVIDIA.

Recent advances in artificial intelligence (AI) and machine learning (ML) offer promising strategies to accelerate protein engineering, yet current AI-assisted protein engineering approaches struggle with limited discovery efficiency and unsatisfactory success rates. In this talk, I will present our ongoing and recent works on the development of large language models (LLMs) and generative AI algorithms for guiding protein engineering from three perspectives — representation learning, prediction, and optimization. First, I will present MSRep, a protein representation learning algorithm that improves protein function annotation. Next, I will introduce ConFit, an efficient algorithm for reprogramming protein LLM to learn the protein sequence-fitness landscape with scarce data, enabling small-data protein engineering. Finally, I will describe MODIFY, an AI-guided algorithm for starting library design in enzyme engineering.

Yunan Luo is an Assistant Professor in the School of Computational Science and Engineering at Georgia Institute of Technology. He is broadly interested in computational biology and machine learning, with a focus on developing AI methods to reveal core scientific insights into biology and medicine. His research has been recognized with several faculty awards, including the NSF CAREER Award and the NIH MIRA Award.

I will describe our efforts in machine learning in chemistry using experimental and computational data sets and what I have learned along the way. Starting from preliminary successes over the past ten years, I will introduce some common questions that arise, such as whether machine learning models can be used to identify any unexpected properties. I will also discuss the tension between known transition metal complex structures and structures that might be anticipated in solution, such as in active catalysts. I will describe how we’ve used experimental data sets to build machine learning models that predict the structure of transition metal complexes, including those that might be present in solution. I will also describe how we have leveraged large datasets of synthesized materials to uncover those with novel function in polymer networks that were unexpected. I will demonstrate the success of our design strategy through macroscopically visible changes in network scale properties of polymers once our transition metal complexes are incorporated.

Professor Heather J. Kulik is the Lammot du Pont (1901) professor in the Departments of Chemical Engineering and Chemistry at MIT. She received her B.E. in Chemical Engineering from the Cooper Union in 2004 and her Ph.D. from the Department of Materials Science and Engineering at MIT in 2009. She completed postdoctoral training at Lawrence Livermore and Stanford, prior to joining MIT as a faculty member in November 2013. Her research has been recognized by an Office of Naval Research Young Investigator Award, DARPA Young Faculty Award and Director’s fellowship, NSF CAREER Award, a Sloan Fellowship in chemistry, an AIChE Computational and Molecular Simulation Engineering Forum Impact Award, a Hans Fischer Senior Fellowship from the Technical University of Munich, and a Presidential Early Career Award for Scientist and Engineers, among others.

Scott E. Denmark was born in New York on 17 June 1953. He obtained an S. B. degree from M.I.T. in 1975 and a D. Sc. Tech degree in 1980 at the ETH-Zürich under the direction of Professor Albert Eschenmoser. That same year he began his career as assistant professor at the University of Illinois. He was promoted to associate professor in 1986, full professor in 1987 and then in 1991 named the Reynold C. Fuson Professor of Chemistry.
His research is primarily focused on the invention of new synthetic reactions and elucidating the mechanisms and origins of stereocontrol in novel, asymmetric reactions. In recent years, his group has investigated the use of chemoinformatics and machine learning to identify and optimize enantioselective catalysts and reaction conditions for a variety of organic and organometallic transformations.
He has been recognized by many awards including the Pedler and Robert Robinson Medals (RSC), the ACS Award for Creative Work in Synthetic Organic Chemistry, the Prelog Medal (ETH-Zürich), the Kipping Award for Research in Silicon Chemistry (ACS), the Ryoji Noyori Prize (Society of Synthetic Organic Chemistry of Japan), the Paracelsus Prize (Swiss Chemical Society) and the Arthur C. Cope Award (ACS). He is a fellow of the American Academy of Arts and Sciences and a member of the US National Academy of Sciences. He edited Volume 85 of Organic Syntheses and serves on the Board of Directors and was Editor in Chief and President of Organic Reactions, Inc. and remains on the Board of Directors.

Molecules built primarily from carbon–carbon bonds, dubbed “small molecules” can do so much good. But most of the functional potential that this class of chemical matter represents remains untapped. Innovation in this space is bottlenecked by lack of access to non-specialists. In fact, all the people in the world that can make small molecules currently fit in one building. The problem is that making carbon–carbon bonds is hard, and doing it repeatedly on a robot is even harder. Blocc chemistry is an emerging solution to this problem. This talk will describe the enabling discovery that MIDA and TIDA ligands can reversibly attenuate the reactivity of boronic acids and thereby enable iterative carbon–carbon bond formation in a manner that is friendly to robots, AI, and anyone. Crucial to achieving automation was the additional discovery that MIDA/TIDA boronates have general binary elution properties on silica gel—enabling an automation-friendly, universal purification strategy after each coupling step. Rather than trying to create machines that can do traditional artisanal chemistry, the Burke group discovered chemistry that machines can do. MIDA/TIDA boronates have now been used by hundreds of academic and industrial labs worldwide to synthesize many different natural products, pharmaceuticals, diagnostic probes, catalysts, quantum dots, organophotovoltaics, and other materials, yielding more than 1000 publications—including more than 300 patent applications. Hundreds of MIDA/TIDA boronates are commercially available. In Burke’s lab, blocc chemistry has been leveraged to develop molecular prosthetics – small molecules that autonomously perform protein-like functions and thus have the potential to treat a wide range of currently incurable diseases, including cystic fibrosis (a first generation molecular prosthetic for CF recently entered the clinic), anemias, and neurodegenerative disorders, as well as a renal sparing antifungal agent which is now in Phase 2 clinical trials. Recent advances in stereospecific Csp3 cross-coupling are substantially expanding the scope of complex small molecule natural products and natural product-like compounds that are accessible with blocc chemistry. Interfacing this approach with frontier AI and automated functional testing methods has enabled closed-loop discovery of new molecular functions. Leveraging the blocc chemistry approach, a Molecule Maker Lab has been created for democratizing molecular innovation.

Martin D. Burke completed his undergraduate studies in Chemistry at Johns Hopkins University, a PhD in Chemistry at Harvard University, and an M.D. in the Health Sciences and Technology Program at Harvard Medical School and Massachusetts Institute of Technology. He is now the May and Ving Lee Professor for Chemical Innovation at UIUC, Founding Director of the Molecule Maker Lab, and co-Founder of the Molecule Maker Lab Institute. He helped launch the Carle Illinois College of Medicine and served as its inaugural Associate Dean for Research.

Burke pioneered block-based chemistry that machines can do. His lab specifically developed MIDA/TIDA boronates to reversibly attenuate the reactivity of boronic acids and thus enable block-based small molecule synthesis via iterative C–C bond formation. He also discovered that MIDA/TIDA boronates share binary elution properties on silica gel enabling generalized catch-and-release purification and thus automation of block-based chemistry. He found that TIDA boronates are >1000X more stable than their MIDA counterparts, powerfully expanding block-based chemistry to include iterative stereospecific Csp3-C bond formation with chiral non-racemic Csp3-rich building blocks, and to accelerate automated block-based total syntheses from 3 days to just 3 hours. He also found that block-based chemistry can be integrated with AI to first achieve closed-loop learning in organic chemistry.

More than 300 of Burke’s building blocks are now commercially available, and they have been used by hundreds of other labs worldwide to synthesize many different types of natural products, pharmaceuticals, herbicides, pesticides, fungicides, diagnostic probes, catalysts, anti-corrosive coatings, quantum dots, carbohydrate sensors, and a wide range of materials, collectively yielding >1000 publications including >300 patents. In his own lab, Burke leveraged this block-based chemistry approach to develop the field of molecular prosthetics, yielding new drug candidates for cystic fibrosis (advanced to clinical trials) and anemia, define the sterol sponge mechanism by which glycosylated polyene macrolide natural products kill eukaryotic cells, which led to renal sparing antifungal candidates for treating invasive fungal infections (advanced to clinical trials), and to enable AI-guided closed-loop discovery of top-in-class organic lasers and mechanistic insights underlying the stability of organophotovoltaic materials.

Burke (co)-founded five biotechnology companies, including REVOLUTION Medicines, Sfunga Therapeutics (now Elion Therapeutics), and cystetic Medicines, that have collectively advanced 7 drug candidates into clinical trials. Burke founded the Molecule Maker Lab (moleculemakerlab.org) and co-founded the Molecule Maker Lab Institute (moleculemaker.org), which are bringing the power of small molecule innovation to anyone. Burke co-created the Digital Molecule Maker, a non-specialist-friendly digital platform for block-based assembly of small molecules that can be “3D-printed” using robots at the MML Burke also helped create a new block-based chemistry curriculum that has already engaged >10,000 students at the undergraduate students, graduate, and K-12 levels.

Burke is an elected member of the National Academy of Medicine and American Society for Clinical Investigation, and a Fellow of the American Association for the Advancement of Science. He is also a winner of the ACS Cope Scholar Award, Elias J. Corey Award in Organic Synthesis, Hirata Gold Medal, Mukaiyama Award, Presidential Medallion from the University of Illinois, and Nobel Laureate Signature Award for Graduate Education in Chemistry. He has also been repeatedly recognized as a Teacher Ranked as Excellent by the UIUC Center for Teaching Excellence.

Modern materials innovation increasingly depends on complex formulations, multi component systems where application performance emerges from coupled effects across composition, microstructure, interfaces, and in some cases chemical reactivity. Navigating these design spaces experimentally is slow and resource intensive, while purely physics based simulations can be limited by scale, sampling cost, and imperfect classical models. In this talk, I will describe an integrated strategy that unifies physics based molecular modeling with machine learning to accelerate formulation design while retaining a mechanistic link to underlying chemistry.

First, I will discuss machine learning models trained on experimental and simulation data to predict formulation properties and quantify trade offs across multiple targets, such as transport, thermomechanical behavior, stability, and interfacial response. Second, I will show how generative and optimization methods can propose new candidate formulations by suggesting promising components and compositions under practical constraints, enabling rapid iteration in large chemical spaces. Third, I will highlight how machine learned molecular potentials can extend molecular dynamics to chemically challenging regimes, including ionic and polar environments, and heterogeneous interfaces, providing higher fidelity simulation data and deeper interpretability. Together, these capabilities support a closed loop workflow for screening, designing, and refining complex materials formulations with improved speed, robustness, and decision making confidence.

Mohammad Atif Faiz (Atif) Afzal is a Principal Scientist at Schrödinger, where he works in the Materials Science division to advance molecular simulation tools and deliver scientific solutions to industrial partners. He earned his undergraduate degree from the Indian Institute of Technology Kanpur and a Ph.D. in Chemical Engineering from the University at Buffalo. His expertise spans molecular dynamics, quantum chemistry, and machine learning for materials modeling and design. Since joining Schrödinger in 2018, Atif has played a key role in collaborating with industry to tackle complex materials challenges, leveraging AI-driven approaches to accelerate innovation.

Agentic AI systems that reason, plan, and act across molecule discovery workflows represent a paradigm shift in how pharmaceutical research operates. This talk presents an architecture integrating generative AI, laboratory automation, and institutional knowledge into a unified platform that eliminates friction from data silos and fragmented tools. We explore how embedded agents automate and parallelize discovery workflows, freeing scientists from sequential bottlenecks and busywork that consume attention. The discussion covers practical lessons in building agent-based systems for therapeutic research, including orchestration patterns and integration with wet lab automation.

Pushkar Ghanekar leads the Frontier AI group within Lilly Small Molecule Discovery, a dedicated team focused on developing and deploying advanced AI systems to accelerate the molecule design and discovery cycle. He earned his Ph.D. in Chemical Engineering from Purdue University, where his research centered on advancing a molecular-level understanding of materials critical for energy storage and conversion to unravel the complex interactions. Outside of work, Pushkar enjoys road biking and any activity that keeps him away from doing laundry and dishes.

Research in the Zahrt group focuses on creating new tools to advance organic synthesis by integrating automation and machine learning workflows to enhance molecular function, reaction efficiency, and sustainability. We develop and use active learning strategies for catalyst design, reaction conditions, and other molecular properties. To accelerate the implementation of these algorithms, we pair them with automated experimentation platforms. Applications of these approaches include exploration of new domains for reaction optimization, new routines for molecular design, and on-demand synthesis of molecules for biological or catalytic applications.

Andrew was born and raised in Fremont, MI, a small rural town in west Michigan. He graduated from Aquinas College (Grand Rapids, MI) in 2014 with degrees in Biology and Chemistry. Later that year, he began his PhD research with Prof. Scott Denmark at the University of Illinois at Urbana-Champaign. As a graduate student, he developed an interest in using computational methods to guide experimental efforts. During this time, he contributed to the development of a data-driven workflow for catalyst design and used applied quantum chemistry for mechanistic elucidation. With this experience working at the interface of experimental chemistry and data science, Andrew became interested in using automated experimentation to streamline implementation of data-driven methods in organic chemistry. This led him to pursue a postdoctoral position in the the laboratory of Prof. Klavs Jensen at the Massachusetts Institute of Technology in 2020. As a member of the Jensen laboratory, Andrew developed a machine-learning-guided workflow for reaction discovery, using it in conjunction with automated experimentation to discover unreported synthetic electrochemical reactions. Andrew’s current interests include developing accessible computational methods to aid experimentalists. This includes developing new data-driven approaches for reaction optimization, catalyst design, analysis, and reaction discovery.

AI is rapidly expanding the toolkit for materials discovery. This talk explores how recent developments in material informatics and large language models have helped unlock new paradigms for materials design and knowledge integration at P&G.

Ajay Muralidharan is currently a Senior Scientist in R&D at Procter & Gamble, where he leverages machine learning and materials modeling to facilitate the discovery of small molecules, surfactants, emulsions, and polymers for formulations. His work emphasizes the integration of machine learning with advanced computational techniques to drive innovation in consumer products. Before joining Procter & Gamble, Ajay earned his PhD from Tulane University and served as a postdoctoral fellow at the University of Wisconsin-Madison, specializing in molecular simulation and the theoretical study of battery materials.

Closed-loop discovery of functional molecules represents a new paradigm to new materials discovery. However, so far, AI-driven closed-loop discovery approaches have rarely yielded new chemical insight. We are addressing this challenge under the Molecule Maker Lab Institute. We connect autonomous synthesis, testing, and machine learning in a closed loop to drive molecular design of light harvesting small molecules towards high photostability. We combined blind Bayesian Optimization (BO) with physics-based ML models to expand the classical “closed loop optimization” method for a more informative “Closed-Loop Transfer” (CLT) paradigm capable of making scientific discoveries. We applied this strategy to a diverse chemical space of 2,200 molecules and discovered exceptional light-harvesting candidates while synthesizing <1.5% of the total chemical space. We further uncovered a first principles understanding of the molecular determinants of photostability in which the triplet excited state manifold plays a critical role. Collectively, these results demonstrate that closed-loop discovery processes can efficiently lead to new fundamental chemical insights, while serving as a practical means for discovering efficient and long-lasting light-harvesting chemistries. We further launched a Kaggle competition to crowd source better ML models to predict photostability. At present, we are working to partner agentic AI with our closed-loop transfer approach for discovery of next generation optoelectronic molecules.

Ying Diao is a Professor, University Scholar, LAS Dean’s Distinguished Professorial Scholar and Dow Chemical Company Faculty Scholar in Department of Chemical and Biomolecular Engineering at University of Illinois at Urbana-Champaign. She serves as the Co-Chair of Molecular Science and Engineering in the Beckman Institute of Advanced Science and Technology, and a Thrust Lead of the Molecular Maker Lab Institute – an NSF AI Institute. She received her Ph.D. degree in Chemical Engineering from MIT in 2012. Her doctoral thesis was on understanding heterogeneous nucleation of pharmaceuticals by designing polymeric substrates. In her subsequent postdoctoral training at Stanford University, she pursued research in the thriving field of printed electronics. Diao group, started in 2015 at Illinois, focuses on understanding assembly of organic functional materials and innovating printing approaches that enable structural control down to the molecular and nanoscale. She has over 130 publications which have been cited >12,000 times. Her work has been frequently featured in scientific journals and news media. She is named to the MIT Technology Review’s annual list of Innovators Under 35 as a pioneer in nanotechnology and materials. She is also a recipient of NSF CAREER Award, NASA Early Career Faculty Award, 3M Non-Tenured Faculty Award, AIChE Allan P. Colburn Award, AIChE Owens Corning Early Career Award and was selected as a Sloan Research Fellow in Chemistry as one of the “very best scientific minds working today”. In 2025, she received the Presidential Early Career Awards for Scientists and Engineers from President Biden.

Abstract:
The discovery and synthesis of functional molecules—ranging from therapeutics to advanced materials—remains a bottleneck in addressing major societal needs. To address this challenge, the Molecule Maker Lab Institute (MMLI) has developed AlphaSynthesis, a comprehensive suite of web-based open-source AI tools and open-access databases designed to accelerate molecular innovation and lower barriers to entry for non-specialists. By uniting these tools into a single platform, AlphaSynthesis enables users to explore prior knowledge, navigate complex chemical space, and predict viable synthetic routes. This presentation introduces AlphaSynthesis platform and showcases key components of the suite through case studies.

Matt Berry is the head of Visual Analytics at the National Center for Supercomputing Applications and has served as Lead Research Software Engineer at the Molecule Maker Lab Institute and on many major grant- and industry-funded initiatives. His primary interest is the creation of user-friendly, production-grade software systems to help knowledge workers understand and act upon complex data.

Abstract:
The fusion of Artificial Intelligence (AI) and chemistry offers unprecedented opportunities for accelerated molecular discovery and synthesis optimization. However, realizing this potential requires an entirely new workforce capable of operating at the intersection of both fields—a critical skill gap currently hindering progress. The Molecule Maker Lab Institute (MMLI) addresses this need through the MMLI Advancing Talent and Research through Interdisciplinary Excellence in AI and Chemistry (MATRIX) program.

The MATRIX program establishes a scalable research training framework designed to democratize access to AI+Chemistry research and expertise. Our mission is to advance talent and foster professional growth across all career stages, from early-career researchers to established PIs, inviting them to become active contributors within the NSF-MMLI community. We achieve this by providing adaptable research training and workshops that bridge the divide between chemical synthesis and computational data science. By lowering technical barriers and promoting collaborative innovation, MATRIX ensures research excellence and equips a diverse generation of molecular innovators to tackle major societal challenges.

Matt Berry is the head of Visual Analytics at the National Center for Supercomputing Applications and has served as Lead Research Software Engineer at the Molecule Maker Lab Institute and on many major grant- and industry-funded initiatives. His primary interest is the creation of user-friendly, production-grade software systems to help knowledge workers understand and act upon complex data.

Authors:
Saif Williams**, James Planey, Anvi Sumant Khot, Shivnandan Subodh Gadgil, Akshaj Manoj Joshi, Swarup Tirupati Iltapawar, Matthew Berry, Chieh-Kai Chan, Sarthak Chandarana, Nolan Green,  Mukund Chorghade, Uday Ohale, Rama Kulkarni, Aarti Khadilkar, Sasikumar Menon, Sucheta Gaikwad, Ruby Mendenhall, Ramona Rudzinski, Agnieszka Lewandowska, Rachel Switzky, and Martin D. Burke

Abstract:
Synthetic organic chemistry has traditionally been an artisanal, manual craft practiced by highly trained specialists. The Molecule Maker Lab (MML) and the Molecule Maker Lab Institute (MMLI) have partnered to democratize molecular innovation by leveraging blocc chemistry—a modular platform for iterative carbon-carbon bond formation based on MIDA/TIDA boronates that is friendly to robots, AI, and non-specialists. Here, we introduce Minecraft for Molecules, a modified version of the Minecraft game that was created by Chicago high school student Saif Williams. This platform enables users to build small molecules within a familiar 3D environment custom built with a molecule reference wall and tutorial signage. A key feature of this platform is the unique capacity to “3D print” molecules directly from the game via a digital-physical connection to MML’s blocc chemistry-based synthesis machines. During the MML ribbon cutting ceremony on June 13, 2025, high school students in Pune, India successfully used Minecraft for Molecules to 3D print a small molecule in real time at the MML in Urbana, USA. This platform enables remote, interactive, and democratized molecular creation—bringing the tools of synthetic organic chemistry to a broad range of future molecular innovators from anywhere in the world.  

Saif Williams is currently a high school student who will be starting his undergraduate degree at UIC this coming Fall. He is NSF-MMLI’s inaugural MATRIX-Edu fellow who has bridged the gap between gaming and automated chemistry. He created a platform that has the unique capacity to “3D print” molecules directly from Minecraft game via a digital-physical connection to MML’s blocc chemistry-based synthesis machines.

Rachel Switzky is the inaugural director of the Siebel Center for Design at the University of Illinois at Urbana-Champaign, a position she has held since 2018. The Siebel Center for Design is dedicated to practicing, modeling, and teaching design thinking, leveraging human-centered design principles to reimagine the campus, community, and the world at large. She also leads the education and outreach work at NSF-MMLI.

Before assuming her current role, Rachel spent over two decades as a global design leader, collaborating with Fortune 100 companies. Most recently, she served as an Executive Director at IDEO, the company that pioneered the concept of design thinking. Throughout the last decade in this capacity, Rachel aided teams in envisioning futures and translating those visions into tangible actions, with a particular emphasis on digital design, emerging technologies, and achieving impact at scale

Burke pioneered blocc chemistry—iterative carbon–carbon bond formation that is friendly to machines, AI, and anyone. His lab developed MIDA and TIDA boronates to reversibly control boronic acid reactivity, enabling iterative C–C bond formation and universal catch-and-release purification. The discovery that TIDA boronates are more than 1,000-fold more stable than their MIDA counterparts expanded blocc chemistry to stereospecific Csp³–C bond formation and accelerated automated synthesis from days to hours.

His team also integrated blocc chemistry with AI, achieving the first example of closed-loop learning in organic synthesis. More than 300 of Burke’s molecular blocc building blocks are now commercially available and have been used by hundreds of laboratories worldwide to synthesize pharmaceuticals, agrochemicals, and advanced materials—yielding more than 1,000 publications and 300 patents.

In his own lab, Burke has applied blocc chemistry to develop molecular prosthetics, generating new therapeutic candidates for cystic fibrosis and anemia, discovering renal-sparing antifungals, and driving AI-guided discovery of novel organic materials.

Justin S. Smith received a PhD degree in computational chemistry from the University of Florida. Smith is known for his pioneering work in the development of machine learning interatomic potentials (MLIPs). As a primary developer of the ANI class of MLIPs, he has significantly advanced the field of computational chemistry by creating models that accurately predict molecular properties and dynamics. In addition to his work on the ANI class of models, he contributed to the development of the AIMNet and HIP-NN family of models. His efforts in these areas have focused on improving the accuracy and efficiency of MLIPs, enabling more precise simulations of molecular systems. He has also been instrumental in advancing active learning techniques for robust data set generation, which are crucial for training accurate and transferable MLIPs. In recent years, Smith has been managing NVIDIA’s strategy in the AI for chemistry and materials science domain. In this role, he has been responsible for overseeing partnerships, prioritization, development and deployment relating to AI-driven tools and technologies that are transforming molecular and materials simulation across domains.

Dr. Mukund Chorghade is a serial entrepreneur and Founder, President, and Chief Scientific Officer, THINQ Pharma and Ayurvidya Healthcare Innovations. He was elected Foreign Fellow National Academy of Sciences, India, in 2023 and was awarded a D.Sc. by the University of Mumbai in 2021. He earned B. Sc. / M. Sc. degrees from the University of Pune, India, and a Ph. D. from Georgetown University. After doing postdoctoral research at the University of Virginia and Harvard University, he directed research groups at Dow, Abbott, CytoMed and Genzyme. He holds/held Chief Research Advisor/ Adjunct Research Professor/Visiting Fellow/ Visiting Scientist appointments at Caltech, Harvard, MIT, Northeastern, Northwestern, Princeton, Rutgers, Univ. of Chicago, School of Medicine-University of Illinois Urbana-Champaign (USA), University of British Columbia (Canada), Cambridge, Leeds, Strathclyde, (UK), College de France, Universite’ Louis Pasteur (France), Universities of Mumbai, Poona, ICT, CSIR, KHRC, Dr. Reddy’s Institute of Life Sciences and SOA (India). He is Chief Scientific Advisor, Late Prin. B.V. Bhide Foundation; Advisor APT Research Foundation Chair, International Advisory Committee-Health, and Bio Sector, DY PATIL International University, Pune and is on Scientific Advisory Board of corporations/foundations such as BVG Life Sciences, Empiriko, YewSavin, World Innovation, Health Sciences Collegium. He is a Faculty of Eminence on the International Advisory Board of Studies at Datta Meghe Institute of Higher Education and Research. He officiates for the Molecular Maker Laboratory Institute External Advisory Board, University of Illinois, Urbana-Champaign. Elected Fellow of the Maharashtra, Andhra Pradesh, and Telangana Academies of Sciences and a recipient of three Scientist of the Year awards, he is a featured speaker in international symposia and serves on Editorial Advisory Boards of Journals. He was elected an ACS Fellow, Section Chair of Brazoria (1990), Northeastern (2007) and Princeton (2019). He participates in ACS’ Career Services/Professional Development/Entrepreneurship and Small Chemicals Businesses.

Andrew is cofounder and head of science at FutureHouse, where he has led multiple AI for science projects at FutureHouse including ChemCrow (first LLM agents in chemistry), ether0 (first reasoning model in a scientific domain), and paperqa (first superhuman literature agent). Andrew has been a professor and researcher in ML in chemistry, explainable AI, statistical mechanics, and chemical engineering, and has received numerous awards, including junior investigator awards from the NSF and NIH.