We’re excited to announce the 2026 Bakar Fellow Award recipients. With the selection of our 15th annual cohort, the Bakar Fellows Program has now surpassed 100 awards and 100 fellows — a milestone that reflects the breadth of researchers and projects we’ve supported over 15 years, and points the way toward supporting many more.
This year’s recipients embody the values at the heart of UC Berkeley’s entrepreneurial ecosystem: innovation, excellence, and collaborative spirit. And though our awardees now number in the hundreds, these fellows were selected for the same reason as every cohort before them. They’ve demonstrated readiness to take on some of the world’s greatest challenges, with the research to back it up.
Since its founding in 2012, the Bakar Fellows Program has worked to expand access to entrepreneurship by providing funding, resources, and mentorship to faculty, graduate students, and undergraduates. In 2023, the program expanded to support partnerships between UC Berkeley and UC San Francisco faculty, helping transform cutting-edge multidisciplinary applications into market-ready solutions.
Two years ago, the program began a key partnership with the Academic Innovation Catalyst (AIC), a deep tech funding platform with a special focus on supporting deep-tech and climate-related projects. What started as a pilot program has now become a multiyear commitment. This exciting expansion strengthens our partnership and will support two new AIC-Bakar Fellows awards each year for the next five years.
Learn more about the 2026 Bakar and AIC-Bakar awardees below, with projects ranging from oncology therapeutics to arsenic removal systems.
Bio-Related Projects
Bakar Fellows – UC Berkeley
Matthew Francis and David Raulet, Efficient Delivery of STING Pathway Agonists Using Targeted Genome-Free Viral Capsids
The “STING” (simulator of interferon genes) pathway is integral to modern-day oncology treatment, due to its ability to trigger immune cells that eliminate tumor tissue. However, the drugs made to activate this pathway have significant limitations that reduce their effectiveness. Francis and Raulet are looking at the STING pathway in a new light by developing a novel delivery platform for its activation based on MS2 viral capsids. This platform will allow drug molecules to be transported in hollow protein shells, shielded from enzymes that would otherwise destroy them. This new method for targeting the STING pathway has great promise for cancer immunotherapy, with mouse model experiments showing significantly reduced tumor burden.
Meng-meng Fu and Matei Zaharia, Using AI and Transcriptome-Wide RNA Structure Mapping to Design ASOs that Treat Neurological Disease
One of the most promising therapies for neurological diseases is antisense nucleotides (ASOs). However, select ASO trials have failed to work on diseases like Parkinson’s and Huntington’s. By combining AI with biological experimentation, Fu and Zaharia plan to create a scalable ASO design process for RNA therapeutics. They will generate a large RNA database, then use bioinformatics and AI to solve RNA structure and design more effective and less toxic ASOs. With Bakar Fellow funding, Fu and Zaharia will develop this cost-effective tool with the potential to equalize accessibility of care for rare neurological diseases.
Teresa Head-Gordon, Mining for Drugs: Exploring Chemical Space with AI Reasoning for Drug Discovery
The development of therapeutics is slow, uncertain, and expensive, with current methods barely scraping the surface of the chemical space where useful compounds may be found. By leveraging AI for chemical discovery, Head-Gordon will explore this chemical space more thoroughly than ever before. She plans to craft a commercially viable discovery engine: AiMiner. Instead of selecting from pre-existing chemical libraries, AiMiner can propose molecules more broadly and then validate them synthetically with LLM’s and high throughput physics validation. Bakar Fellow funding will allow for the conversion of molecular predictions into physical assets, ready to support high-demand fields like oncology.
Bakar Fellows – Berkeley + UCSF Joint Projects
Margaux Pinney and Charles Craik, Engineering a Human IgGase for Transplant Desensitization
Kidney transplantation presents significant challenges, with approximately 30% of patients harboring antibodies that limit donor compatibility and increase graft rejection risk. The bacterial protease IdeS (imlifidase) allows these patients to receive transplants, but can provoke an immune response — limiting its use to once per patient. It also only works on patients without immunity to its bacterial origin, Streptococcus pyogenes (strep throat). Pinney and Craik propose a human-derived enzyme that matches imlifidase’s potency but allows for safe, repeated dosing. This novel enzyme offers exciting new avenues for transplantation and treating autoimmune and antibody-mediated diseases.
Aaron Streets and Diana Laird, mRNA Therapeutics for IVF and Reproductive Longevity
The decline of the egg is one of the most impactful events in human health, yet current science
has accepted egg aging as inevitable. Inspired by studies in fruitflies, Laird discovered that mammalian eggs depend on a stream of precise molecular messages (mRNAs) imported from surrounding cells. With age, imported messages falter alongside egg quality. Amazingly, by resupplying key missing messages, egg function (measured by the potential for healthy embryos) can be restored in middle-aged mice. With Bakar Fellow funding, bioengineer Streets and reproductive scientist Laird plan to translate this discovery into readily deployable advances in fertility preservation for cancer patients, IVF, and developing delivery in vivo to extend ovarian function and protect general health outcomes.
Engineering & Climate-Related Projects
AIC-Bakar Fellows – UC Berkeley
Ashok Gadgil, Cost-effective and Decentralized Arsenic Removal Systems for Small Communities
More than 8 million Americans and over 200 million people globally rely on arsenic-contaminated groundwater as their primary drinking water source. Smaller communities with lower access to crucial infrastructure are disproportionately affected. To combat this problem, Gadgil has developed a 3D-printed, electrochemical-reactor-based technology for treating arsenic-contaminated groundwater. This technology resolves the high operational complexity and cost of its predecessors, making it accessible to smaller communities. Gadgil’s vision is to enable a scalable, economically viable, and community-driven startup for decentralized water treatment using his technology.
Claudio Hail, A High-Power, High-Speed Metasurface Spatial Light Modulator
Fast manipulation of high-power laser beams is an essential technology, used in fields ranging from advanced manufacturing to optical communications. However, existing systems can only steer one laser beam at a time, due to current speed and power limitations. Hail’s work is addressing these shortfalls; he’s developing a device that can shape and direct high-power light more flexibly. With AIC-Bakar funding, Hail’s new device could transform advanced manufacturing, directed energy and telecommunications applications.
Bakar Fellows – UC Berkeley
Alvin Cheung, PandaX: An Agentic Framework for Building Correct, Robust, and
Efficient Data Science Workflows
AI coding tools are now used widely, but there’s a cost: AI-generated code comes without any guarantee of correctness. The alternative is to use standard libraries such as the Python pandas library to write data science pipelines, yet doing so often leads to slow-running code that requires expensive tuning and rewrites. Data scientists end up facing both problems acutely. Cheung’s PandaX framework solves both by integrating directly with the pandas API, using AI to rewrite slow code fragments, find the best implementation for each operation, and prove the rewritten code computes the same results as the original. With Bakar funding, PandaX could reduce data science processing from over a day to minutes while drastically cutting the human effort required to maintain and tune data pipelines.
Asir Intisar Khan, Low-Resistance Topological Semimetals for Energy-Efficient Scaled Interconnects
The rapid growth of high-performance computing and AI is increasing the need for faster, more energy-efficient computer chips. With the shrinking size of computer chips, a growing limitation is the tiny metal wiring, or interconnects, that connect devices inside the chip. Traditional copper interconnects at the nanometer scale become increasingly resistive, leading to slower speeds, increased heat generation, and increased power usage. Khan’s project introduces new materials known as topological semimetal (TSM), a new conductor platform capable of preserving or even improving conductivity as dimensions shrink. Bakar Fellow funding will advance these TSM interconnects from demonstrations all the way to a de-risked, foundry-ready platform that can enable faster, more energy efficient computing technologies.
John Hartwig and Alex Bell, Catalytic Conversion of Waste Polyolefins to Valuable Chemicals
With global plastic pollution forecast to double by 2040, the environmental impact of plastic shows no signs of slowing. Contributing to the problem are two of the largest-volume plastics: polyethylene and polypropylene (PE and PP). These plastics also happen to be the hardest to recycle, with more than 90% of PE and PP ending up in landfills. Hartwig and Bell have developed a type of selective chemical recycling that uses PE and PP to form the high-demand products propylene and isobutylene. These high-volume chemicals can then be used as feedstock for the chemical industry. The potential impacts are powerful: Hartwig and Bell envision a circular carbon economy, where plastic waste becomes a resource instead of a liability.
Ting Xu and Haiyan Huang, Synthetic Enzymes by Materializing Bioinformation
Enzymes are expensive to produce and can easily denature and lose their function, yet they lay claim to a $10B global market and are integral to a greener future. To address this issue, Xu and Huang have developed synthetic replacements to enzymes by translating protein sequence information to random heteropolymers (RHPs). These synthetic enzymes can match a typical enzyme’s function and provide increased stability and reduced cost. The result is a technology that could revolutionize the fields of medical intervention, biotechnology, sustainable manufacturing, and clean energy.
