The Center for Information Technology Research in the Interest of Society and the Banatao Institute (CITRIS) at the University of California (UC) and the UC Berkeley College of Computing, Data Science, and Society (CDSS) are honored to announce the recipients of the 2025 CITRIS-CDSS Innovation Fellowship and AIC Awards.
Made possible with support from the Academic Innovation Catalyst, this program aims to transform faculty-developed research into viable commercial solutions for some of society’s most critical concerns. Each awardee receives up to $200,000 over two years and access to the CITRIS Foundry incubator to help move their work from the laboratory to the market.
“At its core, this program is about advancing technology that serves the public good,” said Matt Sonsini, principal and co-founder of AIC and member of the CITRIS advisory board. “The ingenuity of UC faculty combined with the right support, resources and pathways to scale has the power to turn breakthrough research into solutions that address society’s most pressing challenges.”
This year, more than 85 principal investigators (PIs) applied from CITRIS’s four campuses at Berkeley, Davis, Merced and Santa Cruz. The five winning proposals address deep tech challenges in artificial intelligence (AI) and data science, climate and sustainability, life sciences and biotechnology, and semiconductor design and manufacturing, each focused on translating promising research into solutions with real-world relevance.
“This year’s applicant pool underscores the strength of the University of California as a multi-campus research engine,” said Alexandre Bayen, director of CITRIS and the Banatao Institute, Liao-Cho Innovation Endowed Chair and professor of electrical engineering and computer sciences and of civil and environmental engineering at UC Berkeley, and founding associate provost for the Berkeley Space Center. “The selected proposals highlight how collaboration across Berkeley, Davis, Merced and Santa Cruz is driving new approaches to some of the most complex technical challenges of our time.”
“We are delighted to partner with CITRIS to help bring innovative discoveries that started in UC labs to market,” said Jennifer Chayes, CDSS dean and professor of electrical engineering and computer sciences, mathematics, statistics and information at UC Berkeley. “Thanks to the support of Academic Innovation Catalyst, these awards address a crucial step in the development of new technologies: from idea generation to real-world solutions that benefit society.”
The 2025 award recipients and their projects are:
Cartokit Studio: Visual direct manipulation meets LLM programming
Sarah Chasins, assistant professor of electrical engineering and computer sciences, UC Berkeley
Communicating complex geospatial data remains a major challenge for journalists, civic agencies and researchers, particularly in organizations that lack specialized staff. Existing geographic information system (GIS) tools often force a tradeoff between ease of use and transparency, producing either opaque visual outputs or code-based workflows with steep learning curves.
Cartokit Studio addresses this gap by pairing an intuitive, GIS-style graphical workspace with an AI system that generates and validates the underlying geospatial code in real time. As users edit maps through familiar visual interactions or natural language instructions, Cartokit automatically produces equivalent, explainable Mapbox GL JS programs, keeping code and visual output continuously aligned. This approach enables fast, interactive performance even for large, data-intensive maps.
Designed for newsrooms, city agencies and nonprofits, Cartokit Studio translates cutting-edge programming research into an adoptable, scalable product. Through pilot deployments and an open-core commercialization model, the project aims to lower barriers to trustworthy geospatial analysis and support better informed decision-making across the public interest ecosystem.
A first-in-class treatment for subarachnoid hemorrhage
Theodore Holman, professor of chemistry and biochemistry, UC Santa Cruz
Subarachnoid hemorrhage (SAH) is a severe form of stroke that results in death within 30 days for 25–50 percent of patients, with many survivors experiencing long-term disabilities. Despite its devastating outcomes, effective treatments remain limited.
This research team has developed a novel neuroprotective drug, BPN-27332, which targets the enzyme 15-lipoxygenase (ALOX15), a key driver of early brain injury following SAH. In preclinical studies, inhibiting ALOX15 has been shown to slow disease progression and improve neurological outcomes by up to 85 percent — significantly more effective than the current standard of care. BPN-27332 was developed with strong potency, selectivity and favorable brain penetration, and has demonstrated therapeutic benefit in rodent models.
With this award, the team will compare BPN-27332 to the current standard of care and conduct safety and dose-ranging studies in animal models — critical steps toward advancing this first-in-class therapy to human clinical trials.
FireFly: AI-assisted ground–air wildfire early warning system
Zhaodan Kong, associate professor of mechanical and aerospace engineering, UC Davis
Despite significant investment in satellites and camera networks, more than 80 percent of California wildfires are still first reported by members of the public. Existing optical systems typically detect fires only once smoke or flames are visible, often after containment has become more difficult and dangerous. The core challenge remains early, reliable detection in the first moments of combustion.
FireFly addresses this gap with a low-cost, ground–air wildfire warning system designed for rapid verification. A distributed network of ground-based chemical “tripwire” sensors detects early combustion signatures, such as carbon monoxide and volatile organic compounds. When a potential fire is identified, the system automatically dispatches a long-endurance drone equipped with gas sensors, thermal and optical cameras, and onboard AI to confirm or dismiss the alert within minutes.
The project will develop a full prototype for pilot deployment with the Grass Valley Fire Department in northeastern California, demonstrating a scalable, affordable path to earlier wildfire detection across wildland–urban interface regions.
DNA origami for precision mRNA therapeutics
Greg Tikhomirov, assistant professor of electrical engineering and computer sciences, UC Berkeley
Messenger RNA (mRNA) technology has transformed modern medicine, enabling vaccines and therapies that can be rapidly designed and produced. However, most current mRNA treatments rely on lipid nanoparticles (LNPs) that are difficult to target and can cause unwanted side effects, limiting their effectiveness against diseases such as cancer, HIV, RSV and neurological disorders.
This project introduces a fundamentally new approach to mRNA delivery using RNA–DNA origami nanostructures — compact, programmable carriers built through precise base-pairing of RNA and DNA strands. Unlike conventional delivery systems, these nanoscale structures can be custom-designed for specific tissues, cell types and therapeutic functions, offering precise control over size, shape and targeting.
This project combines advances in nanotechnology, molecular engineering, and synthetic biology to create a scalable platform for next-generation vaccines and gene therapies. By reimagining how genetic medicines are delivered, mRNA origami could expand access to safe and effective therapies worldwide, bringing precision medicine closer to reality for millions of patients.
Desktop electronics projection lithography
Xiaoyu (Rayne) Zheng, associate professor of materials science and engineering, UC Berkeley
Advances in electronics are increasingly moving beyond flat circuits toward fully three-dimensional systems, including curved electronics, embedded electrodes, quantum chips and neural interfaces. However, most manufacturing techniques remain rooted in costly, planar processes such as photolithography, which require cleanroom facilities, multiple materials and long fabrication times. Existing alternatives are often too slow or inflexible to support rapid prototyping or custom geometries.
This project introduces Desktop Electronics Projection Lithography (DEPL), a rapid, single-step fabrication method for producing fully 3D electronic structures. DEPL uses optical patterning and charge-programmed photoresists to precisely control where conductive and insulating regions form, allowing metals to deposit only where needed, without material switching. The approach achieves sub-100-nanometer conductive resolution while dramatically reducing time and cost.
The team will refine DEPL materials, build prototype devices and develop commercialization pathways for applications including robotic sensors, haptic interfaces, antennas and embedded interconnects. The result is a scalable platform that brings complex 3D electronics closer to practical, real-world manufacturing.