The Hertz Foundation is proud to recognize the most recent graduates of the Hertz Fellowship in applied science, mathematics and engineering.
This year’s graduating Hertz fellows have tackled some of the most important scientific and technological challenges of our time. With the flexibility and support of the Hertz Fellowship, they advanced work in viral surveillance, quantum information science, plant and atmospheric biology, neuroscience technologies, materials and energy systems and the mathematical foundations that underpin learning, computation and complex systems.
“These Hertz Fellows are already delivering on their tremendous potential to become leaders in their fields,” said Derek Haseltine, director of the Hertz Fellowship Program. “Their work will continue to strengthen our nation in meaningful and far-reaching ways, from advancing human health to improving our technological and economic competitiveness.”
We congratulate them and look forward to celebrating their accomplishments throughout their careers.
The Hertz Fellowship experience continues beyond the initial award. Graduating fellows join a permanent, multigenerational community that includes some of the nation’s most notable science and technology leaders. The fellowship offers a powerful engine for professional development and collaboration, including lifelong mentoring, events and networking. These connections have helped fellows form new research collaborations, launch companies, commercialize technology and create opportunities together across academia, industry, government and entrepreneurship.
Explore the complete directory of Hertz Fellows to learn more about their diverse backgrounds and interests.
ALLEN LIU
2021 Hertz Fellow
Massachusetts Institute of Technology
Computer Science
Thesis: Learning Theoretic Foundations for Understanding Quantum Systems
Research Summary
Allen Liu uses learning theory to understand how quantum systems behave and how much information we can realistically extract from them. His work centers on a key question in quantum science: what can we efficiently learn about a quantum system, and where do the limits begin?
In his thesis, Liu developed new frameworks for learning the structure and behavior of quantum systems. His research produced several unexpected findings, challenging earlier assumptions about fundamental physical laws and demonstrating efficient algorithms for probing quantum systems in settings once thought to be out of reach. These insights help define the boundary between what is computationally possible in quantum science and what lies beyond.
Liu is currently a Miller Postdoctoral Fellow at University of California, Berkeley and will start as assistant professor in Computer Science at New York University, Courant Institute in fall 2026.
Hertz Fellow Collaborations
Liu collaborated with fellow Hertz Fellow Noah Golowich.
Acknowledgements
Liu would like to acknowledge his Hertz mentor, Ankur Moitra.
“The Hertz Fellowship has provided me with a valuable network of peers and mentors. I also appreciate that having the Hertz Fellowship gives me more independence and freedom to spend time diversifying my research interests and exploring outside my field.”
ALEXANDER COHEN
2022 Hertz Fellow
Massachusetts Institute of Technology
Mathematics
Thesis: Higher dimensional fractal uncertainty
Research Summary
Alexander Cohen’s research explores how waves interact with fractal sets. Waves of different frequencies can either reinforce each other, creating sharp peaks, or cancel out, leaving gaps. In mathematical physics, we often study waves across many length scales, and their constructive interference produces patterns that resemble fractals. Cohen uses properties of fractal sets—such as their tendency to spread out under addition—to analyze wave interference and is particularly interested in how waves spread out on hyperbolic manifolds, an example of a chaotic dynamical system or a setting that natural exhibits chaotic behavior. His work brings together tools from harmonic analysis, fractal geometry, microlocal analysis and quantum chaos to deepen our understanding of wave behavior in complex dynamical systems.
Cohen is now an assistant professor at New York University, Courant Institute.
Graduate Activities + Awards
- Mentor for MIT’s Directed Reading Program (twice)
- Research intern at Microsoft Research New England
- Recipient of a Clay Research Fellowship beginning in 2025
Hertz Fellow Collaborations
Cohen has collaborated with Hertz Fellow Nitya Mani on a published manuscript and with Felipe Hernandez on a project currently in progress.
Acknowledgements
Cohen would like to acknowledge advisor Larry Guth, and mentor Semyon Dyatlov for shaping his research and professional development in harmonic analysis, fractal geometry, microlocal analysis and quantum chaos. He would also like to thank his Hertz mentor, Nitya Mani.
“The summer workshops had a big impact on me. I’m really grateful to have met thoughtful, passionate, and generous people in so many different fields. These workshops have inspired me to think about how my work could connect to other parts of science, and to challenges in the real world.”
CONSTANTINE TZOUANAS
2021 Hertz Fellow
Massachusetts Institute of Technology
Medical Engineering and Medical Physics
Thesis: Rational Engineering of Complex Cellular Ecosystems via Single-Cell Omics
Research Summary
Constantine Tzouanas studies how healthy tissues arise from the properties and interactions of individual cells and how disease develops when these systems break down. His Ph.D. research used single-cell multi-omics and targeted perturbations to examine tissues at unprecedented resolution. By mapping the components, interactions and key regulators that drive tissue function and dysfunction, Tzouanas identified cellular mechanisms behind both healthy physiology and disease. His work shows how engineering approaches and high-dimensional biological data can reveal the rules that govern complex cellular ecosystems.
Tzouanas is a National Science Foundation Graduate Research Fellow and is continuing his research as a Biswas Fellow in the lab of Sangeeta Bhatia at MIT.
Graduate Activities + Awards
- Co-established a mentoring committee in his lab to support junior students during the COVID-19 pandemic
- Interviewed applicants for Ph.D. admissions in his program
- Co-founded high school outreach programs introducing students to underrepresented scientific disciplines
- Co-led global training workshops in South Africa and Thailand to teach single-cell and spatial omics methods
- National Science Foundation Graduate Research Fellow
Acknowledgements
Tzouanas would like to acknowledge his Hertz mentors—David Galas, Michael Loui and Adam Marblestone—for their guidance and support. He also thanks his Ph.D. advisor, Alex Shalek, and the many collaborators who contributed to his work across engineering, biology, chemistry, computation and clinical science.
“The Hertz Fellowship truly embodied its motto of “freedom to innovate”. Thanks to Hertz support, I could explore broadly and ambitiously in order to pursue research directions that would deliver scientific insights and paths to human health impact alike. Hertz further provided a rich, supportive community, connecting me with peers across disciplines in a way that enriched my own thinking and created lifelong professional relationships.”
DAINE DANIELSON
2020 Hertz Fellow
University of Chicago
Physics
Thesis: Gravitationally Mediated Entanglement and Decoherence
Research Summary
Daine Danielson studies some of the deepest questions in modern physics: whether spacetime itself is fundamentally a quantum object. His research shows how upcoming experiments may reveal “gravitationally mediated entanglement,” a phenomenon in which gravity creates quantum connections between massive objects. Detecting such an effect would provide strong evidence for the graviton, the hypothetical quantum particle of spacetime, and would mark the first observation of a phenomenon that cannot be explained by Einstein’s classical theory of general relativity.
Danielson also extended these ideas to extreme astrophysical settings. Working with Robert Wald and Gautam Satishchandran, he discovered that black holes and cosmological horizons act as universal sources of quantum decoherence, gradually destroying the quantum states of nearby systems and forcing them to behave more classically. This fundamental decoherence rate has significant implications for understanding black holes, the structure of the universe and the nature of quantum gravity.
Danielson will continue his research as a Black Hole Initiative Fellow with a joint appointment at the Massachusetts Institute of Technology Center for Theoretical Physics and Harvard University.
Graduate Activities + Awards
- TURIS–IQOQI Fellow at the Institute for Quantum Optics and Quantum Information in Vienna
- Graduate research assistant in the Theoretical Division at Los Alamos National Laboratory, working on inertial confinement fusion and neutrino physics
- Research visits to the Perimeter Institute for Theoretical Physics and the University of Arizona
- Recipient of the APS Blue Apple Prize, a Gravity Research Foundation Essay Award, the University of Chicago Gregor Wentzel Research Prize and the APS Graduate Excellence in Physics Research Award
Acknowledgements
Danielson would like to acknowledge his advisor, Robert Wald, and collaborator Gautam Satishchandran for their influential partnership throughout his dissertation. He also thanks Hertz Fellow Sabrina Pasterski for her support and ongoing collaborations.
“My Hertz Fellowship, the Barbara Ann Canavan Fellowship, has granted me the freedom to travel and form lasting scientific collaborations across continents, and across disciplines within physics, from theory to experiment.”
DOLEV BLUVSTEIN
2019 Hertz Fellow
Harvard University
Experimental Quantum Physics
Thesis: Error-Corrected Quantum Processing With Neutral Atoms
Research Summary
Dolev Bluvstein develops technologies that bring fault-tolerant quantum computing closer to reality. His research focuses on reconfigurable arrays of neutral atoms, a platform that can encode information across many physical qubits and actively correct errors—an essential requirement for building useful quantum computers. Bluvstein’s work helped establish what is now considered a leading approach in the field. His contributions—including new architectures, control methods and demonstrations of logical qubits—provide a foundation for creating the world’s first fault-tolerant quantum processors.
His research played a central role in a landmark publication on a logical quantum processor based on reconfigurable atom arrays, which received the 2024 Physics World Breakthrough of the Year Award.
Bluvstein is a visiting associate in Physics at California Institute of Technology.
Graduate Activities + Awards
- Member of the Center for Ultracold Atoms (CUA) community
- Lead contributor to work recognized as the 2024 Physics World Breakthrough of the Year Award
Acknowledgements
Bluvstein would like to acknowledge Hertz Fellow Jeff Thompson for his mentorship and support. He also thanks his advisor, Mikhail Lukin, whose guidance shaped the direction and impact of his research.
“The friends I met in Hertz empowered me to think big about the impacts of my work, and are quite memorable characters in the adventure of my Ph.D.. It wouldn’t have been the same without them.”
ISAAC METCALF
2020 Hertz Fellow
Rice University
Materials Science
Thesis: Understanding and Suppressing Degradation in Mixed-Dimensional Halide Perovskites
Research Summary
For his Ph.D., Isaac Metcalf worked on understanding the stability of 2D perovskite photovoltaics under different external stimuli and improving their overall performance and stability through surface and interface engineering. The work resulted in a 2024 Science cover featured publication. He will be continuing with Dr. Mohite and the Materials Physics for Energy Management group at Rice University for his Post-doctoral work on further understanding the degradation pathways in 2D perovskites.
Metcalf is a postdoctoral fellow in the Materials Physics for Energy Management group at Rice University.
Graduate Activities + Awards
- Co-first author of a groundbreaking paper on perovskite stabilization, which was featured on the cover of Science in June 2024
Acknowledgements
Metcalf would like to acknowledge the mentors and collaborators who contributed to his work and notes that the independence provided by the Hertz Fellowship enabled him to pursue high-risk research directions that proved both productive and impactful.
JILLIAN PAULL
2020 Hertz Fellow
Harvard University
Systems, Synthetic and Quantitative Biology
Thesis: Integrating Environmental, Clinical, and Metadata Streams for Enhanced Viral Surveillance
Research Summary
Jillian Paull’s research focuses on viral surveillance systems that combine clinical, environmental and metadata streams to deliver clearer, more timely public health insight. During her Ph.D., she helped build systems to monitor SARS-CoV-2 and a broad range of respiratory viruses using both clinical and wastewater sequencing, demonstrating how integrated data can support public health decision-making.
She is continuing her work as a postdoctoral researcher at the Broad Institute, where she is building the next generation of viral surveillance systems. Long term, she hopes to move into an industry role that combines data-focused work with close collaboration across teams.
Graduate Activities + Awards
- Jillian credits starting her family and interning at the Gates Foundation the two most meaningful activities outside of her PhD program.
Acknowledgements
Paull would like to acknowledge her principal investigator, Pardis Sabeti, whose support guided every stage of her Ph.D.. As well as her Hertz mentors: Chris Loose, Philip Welkhoff, Michael Baym, Sarah McFann and Dina Sharon.
“The Hertz community is, to me, the most valuable aspect of being a Hertz Ph.D. Fellow. To know that I have access to a group of highly-qualified, passionate, collaborative scientists has been a huge support.”
JO SCHERRER
2016 Hertz Fellow
Massachusetts Institute of Technology
Neuroscience
Thesis: Driving Temporally Precise Learning in Individual Premotor Neurons using Closed-Loop Neurofeedback
Research Summary
Jo Scherrer develops neurotechnologies that both measure and precisely influence brain activity. During their Ph.D., Scherrer created two complementary tools for understanding and shaping how neural circuits work. The first is a miniaturized implantable microscope that uses optical components inspired by smartphone cameras to record the activity of thousands of neurons at once in freely moving animals. The second is a brain-machine interface built with high-density silicon probes that delivers real-time neurofeedback based on electrical signals from the brain. Using this system, Scherrer demonstrated that the songbird basal ganglia can learn to activate individual cortical neurons at specific moments within a song, offering new insight into how motor learning may operate in similar circuits in humans. These technologies open pathways for highly targeted experiments that test precise circuit-level models of brain function.
Scherrer will continue their work as an inventor at a neurotechnology company.
Graduate Activities + Awards
- Advisory committee member for the Harvard/MGH Interdisciplinary Pain and Headache Rounds
- Contributor to community healthcare and cooperative housing initiatives in Boston
- During their thesis work that involved songbirds learning new motor programs, Jo decided they also needed to learn a new motor program and started learning to play the drum kit, which they now play in a local garage band.
Hertz Fellow Collaborations
Scherrer is a coauthor with Hertz Fellows Sam Rodriques and Ed Boyden on the paper “RNA timestamps identify the age of single molecules in RNA sequencing” (Nature Biotechnology, 2021). They have also collaborated with Dmitriy Aronov on device development for a project in progress.
Acknowledgements
Scherrer would like to acknowledge Hertz mentors Sam Rodriques and Ed Boyden, who supported his early career development at MIT, as well as Philip Welkhoff, Amir Nashat and Daniel Slichter for their continued mentorship throughout their Ph.D. experience. Scherrer would also like to thank their scientific mentors Winfried Denk and Michale Fee for shaping their approach to neuroscience and engineering.
“I have never been in a room with a larger number of simultaneously brilliant, daringly creative, and kind people at once than at Hertz events. The community always challenges me to be the best scientific version of myself, and my work has undoubtedly been more thoughtful and ambitious as a result.”
JOHN CHERIAN
2021 Hertz Fellow
Stanford University
Statistics
Thesis: Conditional Guarantees in Model-Free Inference
Research Summary
John Cherian’s research focuses on making machine learning systems more trustworthy, especially when they are used in high-stakes decisions that affect people and communities. He studied how to evaluate whether a model treats different groups fairly and how confident we can be in a model’s predictions. Many existing techniques appear reliable but often fail to give practitioners the information they actually need.
In his thesis, Cherian approached this challenge from both theoretical and applied perspectives. He developed new methods for assessing fairness and uncertainty in machine learning models and then applied these tools to help build night-of election forecasting models for The Washington Post, showing how stronger statistical foundations can improve real-world decision-making.
Cherian is now a member of the technical staff at OpenAI.
Graduate Activities + Awards
- Collaborated with Isaac Gibbs (now a postdoctoral fellow at UC Berkeley) and Lenny Bronner (at The Washington Post) to design new tools for night-of election forecasting
- Served as a consulting witness in gerrymandering litigation
- Recipient of the Theodore W. Anderson Theory of Statistics Dissertation Award
Acknowledgements
Cherian would like to acknowledge his advisor, Emmanuel Candès, along with collaborators Isaac Gibbs and Lenny Bronner. He also thanks his Hertz mentors Jacob Steinhardt, DJ Strouse, Max Kleiman-Weiner, Adam Jermyn and Anna Bershteyn for their guidance and support.
“The Hertz Fellow community is a constant source of inspiration. It’s easy to feel cynical during a Ph.D. about the value of your research. You’re just one person pushing on what can sometimes feel like an irrelevant and miniscule part of the scientific frontier. Belonging to a community of ambitious but humble fellows has always inspired me to think bigger and more optimistically about my own work. After all, the first step to doing impactful research is simply believing that such a thing is possible to begin with.”
MARISA GAETZ
2020 Hertz Fellow
Massachusetts Institute of Technology
Mathematics
Thesis: Dual Pairs and Disconnected Reductive Groups
Research Summary
Marisa Gaetz’s thesis develops a comprehensive framework for understanding “dual pairs,” which are pairs of subgroups or subalgebras that mutually centralize one another inside larger mathematical structures known as complex reductive Lie groups and Lie algebras. Her thesis provides complete classifications of these dual pairs across all complex classical and exceptional Lie algebras and examines how disconnectedness changes the structure of reductive groups. Using this framework, Gaetz also produced the first full classification of dual pairs in a major nonclassical group, PGL(n,C). Her work advances the broader effort to understand symmetry, structure and representation in high-dimensional algebraic systems.
Gaetz is now an associate officer in disease modeling and health data science at the Gates Foundation’s Institute for Disease Modeling (IDM).
Graduate Activities + Awards
- Co-Founder and President of Brave Behind Bars, Inc.
- MIT Mathematics Award for Service to the Math Community
- George Lusztig PRIMES Mentorship for Exceptional Mentor Service
- Coordinator of MIT’s PRIMES Circle mathematics program for high school students
Acknowledgements
Gaetz would like to acknowledge her Hertz mentor, Philip Welkhoff, whose guidance throughout graduate school was invaluable. She is grateful to continue learning from his leadership at IDM, where his dedication to mentees and to the Hertz community has been a model of generosity and impact.
“While my required experience as a teaching assistant was really positive, it was also quite time consuming. This experience really put into perspective the significant impact that my Hertz fellowship has had on my graduate school experience; I don’t think I could have accomplished nearly as much research without it.”
MAYA SANKAR
2020 Hertz Fellow
Stanford University
Mathematics, Statistics
Thesis: Topological Approaches to Extremal Combinatorics
Research Summary
Maya Sankar’s thesis explores connections between algebraic topology and extremal graph theory and introduces several topologically-inspired tools to measure the global structure of a graph or hypergraph.
Sankar is currently a postdoctoral fellow in mathematics at the Institute for Advanced Study.
Graduate Activities + Awards
- Research Advisor, University of Minnesota Duluth REU
- Mentor, Directed Reading Program, Stanford Math Department
- Board Member & Mentor, Stanford Women in Math Mentoring Program
“The Hertz fellowship’s support towards my Ph.D. has been invaluable throughout my doctoral program. Not having to teach has given me the flexibility to travel freely for conferences and seminars. Such travel has proven crucial for building my network, allowing me to share my results with other mathematicians and, more recently, make an informed decision regarding where I want to be next year.”
MELISSA MAI
2019 Hertz Fellow
Harvard University
Biophysics
Thesis: Architectural Solutions for Hydraulically-Coupled Material Transport in Plants
Research Summary
Melissa Mai explores the biophysical principles that govern how plants function across scales and ecosystems. Her research integrates theoretical modeling with empirical observations, including fieldwork and imaging, to understand the structure–function relationships that drive the coupled transport of fluids and other materials through and within the plant body. She investigates how the architecture of plant tissues shapes processes such as desalination, sugar transport and gas exchange, revealing mechanisms that had previously lacked clear explanation.
In her thesis, Mai identified the material properties that enable salt-tolerant plants to desalinate water using a reverse-osmosis-like process driven by the biomechanics of their waxy cuticles and specialized salt glands. She also demonstrated that the long, narrow geometry of conifer needles is sufficient to allow complex, long-distance sugar transport, resolving a long-standing question in plant physiology. In her final project, she examined the structural determinants of methane emissions from wetland trees. By combining field measurements with exploratory modeling, she identified key physiological and environmental parameters that shape patterns of arboreal methane release and proposed directions for future research in this growing field.
Mai now serves as a AAAS Science and Technology Policy Fellow, sponsored by the American Institute of Physics, in the office of U.S. Sen. Mark Kelly of Arizona.
Graduate Activities + Awards
- Science Education Partner at the Harvard Museum of Natural History, engaging the public through talks, workshops, tours, classroom visits and explainer videos
- Founder and organizer of the Fall Bog Crawl, a community field program exploring the ecology and cultural history of wetlands
- Recipient of Harvard’s Derek C. Bok Award for Excellence in Graduate Student Teaching of Undergraduates
Acknowledgements
Mai would like to acknowledge her principal investigator, Missy Holbrook, for her support throughout the dissertation process, as well as Tomas Bohr, who hosted her at the Technical University of Denmark and played an important role in developing the second chapter of her thesis. She also thanks her Hertz mentor, Brian Camley, for his ongoing encouragement and guidance.
“The Hertz Fellowship and the Hertz community have afforded me the freedom and the support to take risks necessary to cultivate a career that I find fulfilling, meaningful, and impactful.”
NATHANAEL KAZMIERCZAK
2021 Hertz Fellow
California Institute of Technology
Chemistry
Thesis: Illuminating Molecular Spin Relaxation Mechanisms through Ligand Field Theory and Physical Inorganic Spectroscopy
Research Summary
Nathanael Kazmierczak studies how magnetic molecules can be designed to store and process quantum information. Molecules with an unpaired electron spin can act as tiny quantum bits, but random processes that cause the spin to flip destroy the information they hold. In his thesis, Kazmierczak identified the chemical bonding factors that control how quickly these spins reorient. By combining spectroscopy, theory and synthesis, he uncovered strategies for creating molecules whose spins remain stable for much longer periods, offering promise for nanometer-scale quantum sensors and next-generation quantum information technologies.
Kazmierczak is now an assistant professor of chemistry at Hope College in Holland, Michigan, where his lab is building a magnetic circularly polarized luminescence (MCPL) instrument. This technique will support discoveries in photocatalysis, green synthesis, OLED materials and optical molecular quantum bits.
Graduate Activities + Awards
- Teaching assistant for multiple courses, including graduate-level spectroscopy
- Recipient of the Herbert Newby McCoy Award from the Caltech Division of Chemistry and Chemical Engineering
- Integrated spectroscopy, theory and synthesis in his research, an uncommon combination in chemistry
Hertz Fellow Collaborations
Kazmierczak is collaborating with Philip Kocheril on projects involving terahertz spectroscopy and magneto-optical spectroscopy to better understand design principles for molecular quantum bits.
Acknowledgements
Kazmierczak would like to acknowledge Hertz Fellow Cheri Ackerman for inspiring him to apply to the fellowship and for her continued encouragement. He also thanks his advisor, Ryan Hadt, for his mentorship and support throughout the dissertation process.
“The intellectual community is so dynamic. The confluence of fellows in academia, national labs / government, and startups / entrepreneurship gave me a broader perspective on how my scientific research can have an impact on society, and encouraged me to reach for the maximum impact possible.”
NITYA MANI
2019 Hertz Fellow
Massachusetts Institute of Technology
Mathematics, Computer Science
Thesis: A Probabilistic Perspective on Graph Coloring
Research Summary
Nitya Mani’s thesis studies graph coloring from a probabilist’s perspective, focusing on graph coloring problems that share an underlying theme: given an exponentially large family of objects derived from a graph vertex-coloring, can we understand what a typical or random object in this large family looks like without manually searching through exponentially many alternatives? The majority of this thesis is centered around two basic graph coloring problems.
Mani is currently a quantitative trader at Jane Street.
Graduate Activities + Awards
- Emcee for the Advantage Testing Foundation/Jane Street Math Prize for Girls
Hertz Fellow Collaborations
Mani has collaborated with Hertz Fellow Alexander Cohen on a published manuscript.
Acknowledgements
“Thanks to the freedom and support provided by the Fellowship, I was able to spend my final year visiting collaborators as a fellow at SLMath for the special semester in extremal combinatorics.”
NOAH GOLOWICH
2019 Hertz Fellow
Massachusetts Institute of Technology
Computer Science
Thesis: Theoretical Foundations for Learning in Games and Dynamic Environments
Research Summary
Noah Golowich’s research develops theoretical principles behind algorithms which learn and interact efficiently in dynamic environments. His research spans multi-agent learning, reinforcement learning and the mathematics of privacy and trust in machine learning. In multi-agent settings, Golowich analyzed when iterative procedures such as no-regret dynamics converge efficiently to equilibrium, resolving several open problems and establishing near-optimal convergence rates. In reinforcement learning, he characterized the statistical complexity of policy learning and designed algorithms that address challenges such as representation learning, partial observability and interactions among multiple agents. He also developed the first sample-efficient private learning algorithm for general classification and created robust watermarking methods for large language models.
Golowich is currently a postdoctoral researcher at Microsoft Research and will join the University of Texas at Austin as an assistant professor of computer science in Fall 2026.
Graduate Activities + Awards
- Internships at Google Research and Microsoft Research
- Organizer for the MIT Algorithms & Complexity seminar
- Papers and collaborations with leaders across theoretical computer science
- Joining the UT Austin CS faculty in 2026
Hertz Fellow Collaborations
Golowich collaborated with Hertz Fellow Robert Kleinberg on a paper appearing in STOC 2025 and is currently collaborating with fellow Hertz Fellow Allen Liu on projects submitted to conferences.
Acknowledgements
Golowich would like to acknowledge his Hertz mentor and Ph.D. advisor, Ankur Moitra, along with advisors and mentors Constantinos Daskalakis, Sasha Rakhlin, Elad Hazan, Sham Kakade, David Parkes, Madhu Sudan and others who supported his research across visits and internships.
“The Hertz Fellowship has afforded me the freedom to pursue meaningful, impactful, and open-ended research throughout my graduate school experience.”
VIKRAM SUNDAR
2020 Hertz Fellow
Massachusetts Institute of Technology
Computational And Systems Biology
Thesis: Engineering TEV Protease Specificity: An Exploration of Machine Learning and High-Throughput Experimentation for Protein Design
Research Summary
Vikram Sundar works at the intersection of machine learning, structural biology and protein design. With broad interests in applying these tools to drug discovery, his Ph.D. research focused on engineering TEV protease, a sequence-specific protease, to cut new substrates with high specificity. This capability could one day support therapeutic strategies for diseases ranging from cancer to Parkinson’s. To achieve this, Sundar helped develop a high-throughput experimental assay for measuring protease activity and created a Bayesian method for denoising large-scale experimental data, allowing machine learning models to more accurately interpret noisy biological measurements. He then used a dataset of millions of protein variants to train advanced machine learning models and design a new, highly specific TEV protease variant, showing that machine learning can outperform traditional experimental screening alone.
Sundar is currently a machine learning scientist at Generate:Biomedicines, where he works on de novo protein design.
Graduate Activities + Awards
- Classical piano performance as part of MIT’s Emerson/Harris Scholarship Program and Chamber Music Society
- Co-president of the MIT Biotech Group
- Internships at Generate: Biomedicines and Inductive Bio
Acknowledgements
Sundar would like to acknowledge his Hertz mentor and Ph.D. advisor, Kevin Esvelt, as well as his experimental collaborator, Boqiang Tu, for their guidance and support throughout his research.
“The community is welcoming and hugely supportive; it also includes some of the most interesting people I’ve ever met and has educated me on the wildest aspects of science.”