David Baker’s lab at the University of Washington is announcing two major leaps in the field of AI-powered protein design. The first is a souped-up version of its existing RFdiffusion2 tool that can now design enzymes with performance nearly on par with those found in nature. The second is the release of a new, general-purpose version of its model, named RFdiffusion3, which the researchers are calling their most powerful and versatile protein engineering technology to date.

Last year, Baker received the Nobel Prize in Chemistry for his pioneering work in protein science, which includes a deep-learning model called RFdiffusion. The tool allows scientists to design novel proteins that have never existed. These machine-made proteins hold immense promise, from developing medicines for previously untreatable diseases to solving knotty environmental challenges.

Baker leads the UW’s Institute for Protein Design, which released the first version of the core technology in 2023, followed by RFdiffusion2 earlier this year. The second model was fine-tuned for creating enzymes — proteins that orchestrate the transformation of molecules and dramatically speed up chemical reactions.

The latest accomplishments are being shared today in publications in the leading scientific journals Nature and Nature Methods, as well as a preprint last month on bioRxiv.

A better model for enzyme construction

In the improved version of RFdiffusion2, the researchers took a more hands-off approach to guiding the technology, giving it a specific enzymatic task to perform but not specifying other features. Or as the team described it in a press release, the tool produces “blueprints for physical nanomachines that must obey the laws of chemistry and physics to function.”

“You basically let the model have all this space to explore and … you really allow it to search a really wide space and come up with great, great solutions,” said Seth Woodbury, a graduate student in Baker’s lab and author on both papers publishing today.

In addition to UW scientists, researchers from MIT and Switzerland’s ETH Zurich contributed to the work.

The new approach is remarkable for quickly generating higher-performing enzymes. In a test of the tool, it was able to solve 41 out of 41 difficult enzyme design challenges, compared to only 16 for the previous version.

“When we designed enzymes, they’re always an order of magnitude worse than native enzymes that evolution has taken billions of years to find,” said Rohith Krishna, a postdoctoral fellow and lead developer of RFdiffusion2. “This is one of the first times that we’re not one of the best enzymes ever, but we’re in the ballpark of native enzymes.”

The researchers successfully used the model to create proteins calls metallohydrolases, which accelerate difficult reactions using a precisely positioned metal ion and an activated water molecule. The engineered enzymes could have important applications, including the destruction of pollutants.

The promise of rapidly designed catalytic enzymes could unleash wide-ranging applications, Baker said.

“The first problem we really tackled with AI, it was largely therapeutics, making binders to drug targets,” he said. “But now with catalysis, it really opens up sustainability.”

The researchers are also working with the Gates Foundation to figure out lower-cost ways to build what are known as small molecule drugs, which interact with proteins and enzymes inside cells, often by blocking or enhancing their function to effect biological processes.

The most powerful model to date

While RFdiffusion2 is fine-tuned to make enzymes, the Institute for Protein Design researchers were also eager to build a tool with wide-ranging functionality. RFdiffusion3 is that new AI model. It can create proteins that interact with virtually every type of molecule found in cells, including the ability to bind DNA, other proteins and small molecules, in addition to enzyme-related functions.

“We really are excited about building more and more complex systems, so we didn’t want to have bespoke models for each application. We wanted to be able to combine everything into one foundational model,” said Krishna, a lead developer of RFdiffusion3.

Today the team is publicly releasing the code for the new machine learning tool.

“We’re really excited to see what everyone else builds on it,” Krishna said.

And while the steady stream of model upgrades, breakthroughs and publications in top-notch journals seems to continue unabated from the Institute for Protein Design, there are plenty of behind-the-scenes stumbles, Baker said.

“It all sounds beautiful and simple at the end when it’s done,” he said. “But along the way, there’s always the moments when it seems like it won’t work.”

But the researchers keep at it, and so far at least, they keep finding a path forward. And the institute continues minting new graduates and further training postdocs who go on to launch companies or establish their own academic labs.

“I don’t surf, but I sort of feel like we’re riding a wave and it’s just fun,” Baker said. “I mean, it’s so many, so many problems are getting solved. And yeah, it’s really exhilarating, honestly.”

The Nature paper, titled “Computational design of metallohydrolases,” was authored by Donghyo Kim, Seth Woodbury, Woody Ahern, Doug Tischer, Alex Kang, Emily Joyce, Asim Bera, Nikita Hanikel, Saman Salike, Rohith Krishna, Jason Yim, Samuel Pellock, Anna Lauko, Indrek Kalvet, Donald Hilvert and David Baker.

The Nature Methods paper, titled “Atom-level enzyme active site scaffolding using RFdiffusion2,” was authored by Woody Ahern, Jason Yim, Doug Tischer, Saman Salike, Seth Woodbury, Donghyo Kim, Indrek Kalvet, Yakov Kipnis, Brian Coventry, Han Raut Altae-Tran, Magnus Bauer, Regina Barzilay, Tommi Jaakkola, Rohith Krishna and David Baker.

In an unusual act of philanthropy, an anonymous donor has committed more than $50 million to the University of Washington to support the little-known field of medical laboratory science. The funds will be distributed over the next half-century.

UW leaders called the gift “transformational,” noting it’s the largest gift they’re aware of for this particular specialty.

The donation will immediately impact the current class of 35 students in the Medical Laboratory Science Undergraduate Program by covering their tuition costs — waiving about $9,000 per student — during the two quarters of clinical rotations in their senior year.

When the students learned the news at an event Monday at the UW’s Seattle campus, many began to cry.

Students who graduate with a four-year degree in medical laboratory sciences are essential, behind-the-scenes healthcare workers. They collect biological samples, process the material, help interpret the results, and provide necessary data for individual patients and public health institutions.

Dr. Geoff Baird, chair of the Department of Laboratory Medicine and Pathology at UW Medicine, praised the program for training these healthcare professionals.

“No one really ever pays attention to the glue that holds the whole thing together,” Baird said of their critical role.

Dr. Tim Dellit, UW Medicine CEO and the dean of the School of Medicine Tim Dellit, echoed the sentiment in sharing news of the gift with the students. “In many ways, you are the unsung heroes,” he said. “You work behind the scenes that allow all of the healthcare machinery to continue to work.”

The field, however, is facing a challenge. Despite its importance, the workforce is aging, and there aren’t enough students graduating with the needed expertise, said Baird. The new gift is designed to help address that shortage by expanding the two-year medical laboratory sciences program from the current 70 students to 100 over the next decade.

Graduates earn a four-year bachelor’s degree and professional certifications, ready for employment at clinics and hospitals.

The university didn’t share details about the donor, except to say that he is a Washington resident and a big fan of the longtime, local burger franchise, Dick’s Drive-In. To celebrate the news, he requested that the students were served burgers at the announcement.

For the students, the financial relief felt profound.

Senior Lily Koplowitz-Fleming was grateful that she won’t have to juggle an additional job on top of the nine-hours, five-days a week that’s required by the clinical rotation. Instead, she’ll be able to focus on the training for her future career, which she said is a meaningful blend of “skills-based and knowledge-based” work.

Another senior, Keila Uchimura, also said she enrolled in the program because she “really likes being able to see the direct impact you make.”

While medical lab scientists typically work in the background, their roles became more noticeable during the pandemic as people rushed to get tested and waited anxiously for results.

Baird praised the donor and his gift in an earlier GeekWire interview.

“The morality, the righteousness of it — it’s just really impressive that someone was able to find that generosity,” he said. “And we’re all in the state of Washington forever indebted — not just the students.”

The University of Washington is powering up its vision for a brighter, more sustainable future with a newly completed solar canopy installed in a sprawling parking lot north of Husky Stadium.

The 84-kilowatt solar array is paired with Level 2 EV charging that can accommodate 20 vehicles simultaneously. The $3.7 million project includes electrical infrastructure to support the future installation of panels capable of nearly 30 times more power generation — up to 2.5 megawatts. That’s enough capacity to power roughly 2,000 homes.

The solar canopy is a pilot project supporting the UW’s goals to cut its carbon footprint, said Mark Huppert, interim director of UW Transportation Services.

“Located on the site of the former Montlake landfill, the pilot demonstrates how the land can be put to work to achieve more sustainable outcomes,” Huppert said via email.

Project partners include Sea Con as the general contractor and Prime Electric as the electrical contractor. The canopy system was fabricated by Trinity Structures, which has since rebranded as Trinity Energy.

The installation is connected to electrical grids powering the City of Seattle and the UW’s campus. The ability to generate energy onsite can curb the university’s reliance on the utility grid while reducing the impacts of power outages and fluctuating electricity costs.

“Generating solar power from a parking lot may sound modest, but the strategic value is enormous,” said Darin Leonard, president of Trinity Energy, in a statement.

The idea for the project grew out of a collaboration between the student organization UW Solar; Anne Eskridge, the retired director of UW Transportation Services; and Jan Whittington, director of the UW’s Urban Infrastructure Lab.

The university is currently drafting its 2050 Sustainability Action Plan, which includes the long-term expansion of the parking lot solar canopies.

The UW Solar students “will continue to support the efforts to achieve the vision of a complete build-out,” Huppert said.

The project was funded by UW Transportation Services, Seattle City Light and Washington state’s Climate Commitment Act, administered through the Washington State Department of Commerce’s electric vehicle charging program.

Editor’s note: Story updated to provide additional information on project partners.

University of Washington President Robert Jones wants to expand computer science access for undergraduates and build new public-private partnerships to tackle society’s grand challenges — and he has concrete ideas on how to make that happen.

More than 100 days into his tenure as the UW’s 34th president, Jones is also working to dispel two persistent myths: that it’s nearly impossible to get into the UW’s Paul G. Allen School of Computer Science & Engineering, and that AI is taking everyone’s jobs.

This fall, the Allen School — a top tech program nationally — accepted 37% of the direct applicants from Washington state high schools, though out-of-state admissions is only 4%. “We actually accept many more students than the general public believes,” Jones said in an interview this week with GeekWire.

As to the AI job apocalypse? “That’s an overblown fear,” Jones said. AI is a “critically important tool to have in your tool chest to be more effective in the future.”

Drawing on his experience leading the University of Illinois Urbana-Champaign, Jones wants to give a bigger slice of the UW’s 45,000 undergraduates access to computer science courses — even if they’re in unrelated degree programs.

The idea is modeled on the groundbreaking “CS + X” initiative that he helped expand during his past presidency in Illinois. The program created tech-infused studies in 17 degree programs including advertising, astronomy, economics, music, philosophy and physics.

In Illinois, CS + X launched in crop sciences more than a decade ago because agricultural tech was, and continues to be, one of the fastest growing sectors in the state. Jones himself began his career as a professor in plant physiology and became an international authority in the field.

“AI is just an amazing tool, and we’re doing work here to try to make sure that our students are getting as comprehensive an education as possible,” Jones said. Bolstering a graduate’s knowledge of computer science and artificial intelligence in addition to their focus area makes them “much more employable.”

A $10 million gift announced Tuesday from Microsoft pioneer Charles Simonyi and his wife, Lisa Simonyi, will help facilitate the integration of AI into education and research across the university through the newly formed AI@UW initiative.

AI@UW, which includes a new Vice Provost for Artificial Intelligence position, will help the UW maintain its “strategic advantage” as an AI leader, Jones said. It’s also an example of the sort of public-private collaborations he’d like to foster.

“Radical partnerships”

Jones envisions expanding what he calls “radical partnerships” — building diverse coalitions across geographies, institutions and sectors to tackle problems “that are too big for any one entity to solve alone.”

The collaborations can bring together both expertise and funding. The university faces a difficult financial landscape as Washington state’s revenue forecast continues to weaken, drawing down an already strained budget. Add to that ongoing worries about funding cuts to federal research programs that the UW heavily relies upon.

Jones pointed to the university’s long-running WWAMI program, which serves medical students from Washington, Wyoming, Alaska, Montana and Idaho who earn their degrees at the UW and return to their communities to practice medicine, as a prime example of the radical partnership approach.

“I’ve done about three or four of these over the last 10 years,” he said, “and they’re just amazing ways to bring people together and to think about doing research in a different, in a much more collaborative, much more impactful way than we’ve ever thought about.”

The strategy is particularly suited to AI, quantum computing or other massive technological challenges, he said. Jones was involved in partnerships with the University of Chicago to bolster quantum science research in Illinois and a human biology collaboration with the Chan Zuckerberg Biohub Chicago.

In future collaborations in his new role, Jones is eager to work “seamlessly” with the region’s tech sector, leveraging the fact that the UW helped create an ecosystem that includes giants like Amazon and Microsoft, as well as hundreds of smaller companies.

“Given Dr. Jones’s work creating the quantum park in Chicago, I have no doubt that we will continue to see the UW play an integral role in the growth of the ecosystem,” said Laura Ruderman, CEO of the TechAlliance, who also called the UW “critical” to the state’s innovation economy.

More than 110 UW spinoffs currently operate in the state, according to the university’s CoMotion program, which supports entrepreneurship.

“What we have to do more of is working together to fund and to create the next big idea that’s going to be transformative,” Jones said, “not only for the state of Washington, but for the nation and the world.”

Editor’s note: This series profiles six of the Seattle region’s “Uncommon Thinkers”: inventors, scientists, technologists and entrepreneurs transforming industries and driving positive change in the world. They will be recognized Dec. 11 at the GeekWire Gala. Uncommon Thinkers is presented in partnership with Greater Seattle Partners.

After earning his bachelor’s degree in mechanical engineering, Brian Pinkard spent six months “flipping rocks,” as he describes it, in Colorado’s Rocky Mountains.

The rock-flipping was purposeful work: Pinkard was clearing obstructions and building trails for AmeriCorps, spending every night in a tent.

“I loved it. It was great. And the reason I did that is because I wanted to do something that mattered, that made a difference in the world,” he said. When the program ended, he was inspired to direct his impact to a larger environmental challenge.

His passion to do good, paired with an engineer’s drive for problem solving, led him to a doctoral degree from the University of Washington and then to launching Aquagga, a startup that’s destroying PFAS — a toxic class of pollutants known as “forever chemicals.”

“Brian has been very laser focused on his mission,” said Igor Novosselov, Pinkard’s PhD advisor and research professor at the UW’s Department of Mechanical Engineering. “He’s not a typical scientist who would just go and write a bunch of papers. He’s going after impact where it matters.”

But a few steps before PFAS, Pinkard was focused on nerve gas in the Middle East.

‘Nobody knows how to treat this stuff‘

When Pinkard joined Novosselov’s lab, it had U.S. Department of Defense funding to develop an in-the-field, mobile strategy for treating barrels of abandoned chemical weapons in the Syrian desert. The previous solution was to truck the barrels to the Mediterranean Sea, load them on a boat and incinerate the material.

“If you’re the guy who’s got to transport a nerve agent,” Pinkard noted, “it’s not a very good job.”

Within five years, the lab came up with a workable solution, but the need was no longer urgent and DoD shelved its application of the technology, though Novosselov continued to work on it.

Pinkard appreciated the tremendous power of the strategy for treating dangerous materials and wondered if there was another use case. Then as he was preparing to finish his PhD in June 2020, the COVID pandemic hit, derailing his plans to apply for a university postdoctoral fellowship as no one was hiring.

So he made a pivot to entrepreneurship — a role he had never considered.

Pinkard teamed up with engineer and tech innovator Nigel Sharp to explore the potential for using the tech, called supercritical water oxidation, to treat sewage sludge from wastewater treatment plants, but they realized the market wasn’t viable.

There was, however, buzz about PFAS.

“Everybody was talking about PFAS,” he said, and if anyone could figure out how to destroy the chemicals, it would be a breakthrough. That realization became his lightbulb moment.

Destroying PFAS

PFAS is a family of chemicals that for decades have been added to firefighting foams, food packaging, carpets and fabrics, water-repellent clothing and non-stick pans. The resilient chemicals are great at deflecting water, stains and grease — but they escape from products and now contaminate drinking water across the nation and are even in mothers’ breast milk.

PFAS are still in use, while researchers and regulators are increasingly concerned by their serious health impacts.

Pinkard and Sharp launched Aquagga in 2019 in Tacoma, Wash., and were soon joined by co-founder Chris Woodruff. The team kept the idea of modular treatment units but shifted to a related but different chemistry (hydrothermal alkaline treatment) for destroying PFAS, securing a patent for the approach from the Colorado School of Mines.

“Brian has been a great partner from the beginning,” said Timothy Strathmann, a Colorado School of Mines professor. “Unlike many entrepreneurs I’ve interacted with, he is also deeply interested in understanding the limitations and technical challenges associated with the technology. He’s keenly aware that the long-term success of Aquagga will only be achieved by addressing the critical barriers to deployment.”

Aquagga’s devices annihilates PFAS under super hot, high pressure conditions made caustic and corrosive through the addition of lye.

The company has done nine field demonstrations of its technology, including a project at an airport in Alaska, a DoD-funded project in North Carolina involving firefighting foams, and a wastewater demo with the City of Tacoma. It’s now close to signing its first long-term commercial deployment, Pinkard said, “which will be a huge milestone for us.”

“It’s really cool to see how much PFAS we’ve destroyed … even in our short journey,” Pinkard said. “And to think about where it could go, what it could enable at scale. So [I’m] very optimistic about Aquagga’s future. I’m very optimistic about the impact we could create, the lives we could save.”

The University of Washington today announced a $10 million gift from Microsoft pioneer Charles Simonyi and his wife, Lisa Simonyi, to launch AI@UW, a campus-wide initiative supporting the university’s leadership in the responsible, effective use of artificial intelligence in the classroom and research.

The initiative creates a new Vice Provost for Artificial Intelligence position, with Professor Noah Smith of the Paul G. Allen School of Computer Science & Engineering serving in the inaugural role.

Smith said there’s a huge amount of expertise around smart AI adoption at the UW, “and what I really want to do is connect all of that, bring it together, map out what people already know and are doing, cast the light on it so we can all learn from each other more effectively and accelerate it.”

Professors are eager — perhaps even desperate — for support navigating AI’s role in education.

“If you ask faculty what’s the one question on their minds right now, it’s: ‘My students are using AI. What now? What am I supposed to do? How do we respond to this?'” Smith said.

The UW’s response is setting a path where AI assists students by answering questions or prepping study tools, but doesn’t do the work for them. On the faculty side, it can aid in creating fair and useful tests and evaluations. Smith is advocating for conversations and transparency in helping students find the balance where AI complements their academic journey, but doesn’t replace their education.

“You don’t go to university,” he said, “if you don’t actually want to learn.”

A key component of AI@UW is its grant program — SEED-AI, which stands for Supporting Educational Excellence and Discovery. The grants will provide funding to faculty across the university who have innovative, exploratory approaches for using AI in their courses that could be widely adopted. The call for grant proposals should go out in the next few weeks.

Smith highlighted three additional focus areas:

• Governance and policy: Creating a governance committee that establishes infrastructure for setting AI use policies that facilitate innovation in the classroom.
• AI literacy courses: Developing courses for all undergraduates addressing AI literacy from different disciplinary perspectives so students “have an understanding of AI that is not grounded in fear or grounded in fantasy and hype,” Smith said.
• Expert network: Forming a network of AI experts within the UW who can assist faculty working on research and education projects and need a customized AI tool.

UW President Robert Jones said the initiative and new vice provost role will help the university maintain its “strategic advantage” as a leader in AI.

“We need somebody that wakes up each and every day that thinks about AI across the three parts of our mission: our teaching, our research and our innovation agenda,” Jones said in a GeekWire interview. “So that’s the value proposition.”

Including the donation announced today, Charles and Lisa Simonyi have given more than $27.5 million to the UW since 2009, supporting DIRAC (Data Intensive Research in Astrophysics & Cosmology), the Ana Mari Cauce Welcome Center and the Allen School building.

Charles Simonyi, who has a net worth north of $8 billion, was a groundbreaking software architect at Microsoft and remains a technical fellow with the Redmond, Wash.-based company, while Lisa Simonyi is chair of the UW Foundation Board.

The new gift also establishes the Charles and Lisa Simonyi Endowed Chair for Artificial Intelligence and Emerging Technologies, with Smith selected as the inaugural recipient.

And in addition to his role at the Allen School, Smith is also affiliated with the Department of Linguistics, the Center for Statistics and the Social Sciences, the eScience Institute, and the Stroum Center for Jewish Studies, providing useful experience in working across disciplines.

Smith is reaching out to other institutions who are likewise pioneering programs to employ AI on campus to learn about their efforts, but added that the UW has an advantage with the new funding. “The Simonyi gift,” he said, “is going to set us ahead.”

Editor’s note: This series profiles six of the Seattle region’s “Uncommon Thinkers”: inventors, scientists, technologists and entrepreneurs transforming industries and driving positive change in the world. They will be recognized Dec. 11 at the GeekWire Gala. Uncommon Thinkers is presented in partnership with Greater Seattle Partners.

Before he launched a venture-backed biotech startup, prior even to landing a research role in one of the world’s premier academic labs, Anindya Roy arrived in the U.S. with two suitcases and $2,000 in the bank.

Roy grew up in rural India in a home that lacked electricity and running water during his childhood. A passion for science fueled his ambitions, leading him to earn degrees at the University of Calcutta and the Indian Institute of Technology in Kharagpur.

Then he made the bold leap in 2008 to pursue his PhD at Arizona State University, which led to a postdoctoral fellowship with David Baker, a University of Washington professor who last year won a Nobel Prize in Chemistry.

In 2023, Roy co-founded Seattle-based Lila Biologics, which uses the AI-powered protein design technology developed in the Baker lab to pursue cutting-edge medical therapies.

“Anindya is a brilliant and determined scientist and innovator who has made key contributions across diverse areas of science,” Baker said, “and is charting a most exciting path forward with Lila.”

Dr. Sheila Gujrathi, a biotech executive and chair of Lila’s board of directors, described Roy as “a thoughtful and creative problem-solver who approaches each challenge with genuine humility. He stands out not just for his innovative thinking, but also for his sincere kindness and integrity.”

Unlocking potential

In the lab at ASU, Roy focused on protein engineering for sustainable energy resources, but he was eager to apply those skills to medicine. He sent an email to Baker who invited him for an interview and tour of his protein creation lab, which delivered a kid-in-a-candy-shop kind of experience.

“That was the most exciting thing because it was such an amazingly diverse set of computational protein design problems, aiming to solve so many different kinds of things,” Roy recalled.

He jumped at the postdoc opportunity, joining the lab that is part of the UW’s Institute for Protein Design (IPD). There he began exploring the groundbreaking tools for creating proteins from scratch, ultimately pursuing a molecule that showed promise in cancer care and the treatment of fibrotic diseases that form scar tissue in various organs.

Roy eventually entered the IPD’s Translational Investigator Research Program, which gives entrepreneurial scientists the support and training to begin commercializing their discoveries. Two years ago, he and Jake Kraft, a fellow IPD postdoc, licensed the molecule they worked on at the UW and launched Lila.

While Roy has found success in his research, scientific inquiry can be slow-going and frustrating. To unwind he turns to intense weight training and goes to live shows — he caught Lady Gaga this summer and loves house music. Roy also whips up French pastries and tortes worthy of “The Great British Bake Off.”

And sometimes he reflects on the unlikely journey that led him to launching his own company.

“Whenever I get kind of discouraged or depressed about things, I look back at my career trajectory and how far I’ve come,” Roy said. “That does give me a lot of strength.”

The power of science

His startup is also making confidence-boosting progress. Lila has raised $10 million from investors and released two AI-powered platforms for creating therapeutic proteins. One is focused on targeted radiotherapy, generating proteins that precisely bind to tumors and carry radioactive isotopes that zap cancerous cells. The other platform is used to build long-acting injectable drugs that slowly release medicine over weeks or months.

In September, the seven-person startup announced a collaboration with pharmaceutical giant Eli Lilly to develop therapies for treating solid tumors.

Roy is grateful for U.S. support of the basic research that underpins the work being done at universities, institutions and companies nationwide. He’s also worried about federal funding cuts being pursued by the current administration that threaten America’s leadership in scientific innovation.

Because while he has been doing de novo protein design for more than a decade, Roy is still amazed by what the technology can do and how fast it’s evolving.

“This is almost like science fiction,” Roy said. “Years ago, you never imagined what we are doing right now. You are designing molecules in the computer, and you are putting them in actual living systems, and it’s doing what it’s supposed to do. It is pure science fiction.”

Hearvana, a Seattle startup using AI to create “superhuman hearing capabilities,” raised $6 million in pre-seed funding.

The company was launched this spring by University of Washington computer science researchers, including co-founder Shyam Gollakota, a renowned tech inventor.

Gollakota previously told GeekWire that Hearvana is “creating AI breakthroughs that are shaping the future of sound” on-device, using the technology to quickly process audio without requiring large amounts of power or computing.

He predicted the tech would be part of billions of earbuds, hearing aids and smartphones.

In a new LinkedIn post on Wednesday, Gollakota shared more about Hearvana:

“Hearvana AI is a result of years of research and experience creating super-human and proactive audio AI systems. Our platform represents a major leap in real-time and on-device audio augmentation and comprehension, enabling AI assistants, hearing devices, smart glasses, and voice-driven products to listen, interpret, and manipulate audio with unprecedented quality and latency. Hearvana empowers devices to perceive the world beyond human capability, grasping intent and context in complex, noisy environments, and helping people communicate more effectively with each other and with machines.”

A professor at the UW’s Paul G. Allen School of Computer Science & Engineering, Gollakota is head of the Mobile Intelligence Lab. He previously co-founded Sound Life Sciences, a UW spinout that developed an app to monitor breathing that was acquired by Google in 2022. And he’s the co-founder of Wavely Diagnostics, which uses a smartphone app to detect ear infections.

Malek Itani, a research assistant and PhD student at the Allen School, is a co-founder of Hearvana. Itani was an intern at Meta, where he worked on smart glasses.

The pair conducted previous research on a headphone prototype that used AI to create a “sound bubble” in noisy environments and could learn the distance for each sound source in a room. 

Hearvana is being incubated at the AI2 Incubator in Seattle.

The investment round was led by Point72 Ventures and SCB 10X with participation from the AI2 Incubator, SBI US Gateway Fund, Forston VC, Ascend, J4 Ventures, Pack Ventures, Moai Capital, and Amazon Alexa Fund.

Axios first reported on the funding.

Researchers from Nobel Laureate David Baker’s lab and the University of Washington’s Institute for Protein Design (IPD) have used artificial intelligence to design antibodies from scratch — notching another game-changing breakthrough for the scientists and their field of research.

“It was really a grand challenge — a pipe dream,” said Andrew Borst, head of electron microscopy R&D at IPD. Now that they’ve hit the milestone of engineering antibodies that successfully bind to their targets, the research “can go on and it can grow to heights that you can’t imagine right now.”

Borst and his colleagues are publishing their work in the peer-reviewed journal Nature. The development could supercharge the $200 billion antibody drug industry.

Before the advent of AI-based tools, scientists made antibodies by immunizing animals and hoping they would produce useful molecules. The process was laborious and expensive, but tremendously important. Many powerful new drugs for treating cancer and autoimmune diseases are antibody-based, using the proteins to hit specific targets.

Baker, who won the Nobel Prize in Chemistry last year, was recognized for his work unraveling the molecular design of proteins and developing AI-powered tools to rapidly build and test new ones. The technology learns from existing proteins and how they function, then creates designs to solve specific challenges.

In the new research, the team focused on the six loops of protein on the antibody’s arms that serves as fingers that grab its target. Earlier efforts would tweak maybe one of the loops, but the latest technology allows for a much bigger play.

“We are starting totally from scratch — from the loop perspective — so we’re designing all six,” said Robert Ragotte, a postdoctoral researcher at IPD. “But the rest of the antibody, what’s called the framework, that is actually staying the same.”

The hope is that by retaining the familiar humanness of most of the antibody, a patient’s immune system would ignore the drug rather than mount an offense against an otherwise foreign molecule.

The researchers tested their computer creations against multiple real-world targets including hemagglutinin, a protein on flu viruses that allow them to infect host cells; a potent toxin produced by the C. difficile bacteria; and others.

The lab tests showed that in most cases, the new antibodies bound to their targets as the online simulations predicted they would.

“They were binding in the right way with the right shape against the right target at the spot of interest that would potentially be useful for some sort of therapeutic effect,” Borst said. “This was a really incredible result to see.”

Borst added that the computational and wet lab biologists worked closely together, allowing the scientists to refine their digital designs based on what the real-life experiments revealed.

The software used to create the antibodies is freely available on GitHub for anyone to use. Xaira Therapeutics, a well-funded biotech startup led by IPD alumni, has licensed some of the technology for its commercial operations and multiple authors on the Nature paper are currently employed by the company.

While the antibodies created as part of the research demonstrated the software’s potential, there are many more steps to engineering a potential therapy. Candidate drugs need to be optimized for additional features such as high solubility, a strong affinity for a target and minimizing immunogenicity — which is an unwanted immune response.

Before joining IPD four years ago, Ragotte was a graduate student doing conventional antibody discovery and characterization using animals.

The idea that one day you could get on a computer, choose a target, and create a DNA blueprint for building a protein was almost unimaginable, he said. “We would talk about it, but it didn’t even seem like a tractable problem at that point.”

The Nature study is titled “Atomically accurate de novo design of antibodies with RFdiffusion.” The lead authors include Nathaniel Bennett, Joseph Watson, Robert Ragotte, Andrew Borst, DéJenaé See,
Connor Weidle and Riti Biswas, all of whom were affiliated with the UW at the time the research was conducted, and Yutong Yu of the University of California, Irvine. David Baker is the senior author.

Additional authors are: Ellen Shrock, Russell Ault, Philip Leung, Buwei Huang, Inna Goreshnik, John Tam, Kenneth Carr, Benedikt Singer, Cameron Criswell, Basile Wicky, Dionne Vafeados, Mariana Sanchez, Ho Kim, Susana Torres, Sidney Chan, Shirley Sun, Timothy Spear, Yi Sun, Keelan O’Reilly, John Maris, Nikolaos Sgourakis, Roman Melnyk and Chang Liu.

The University of Washington’s Paul G. Allen School of Computer Science & Engineering is reframing what it means for its research to change the world.

In unveiling six “Grand Challenges” at its annual Research Showcase and Open House in Seattle on Wednesday, the Allen School’s leaders described a blueprint for technology that protects privacy, supports mental health, broadens accessibility, earns public trust, and sustains people and the planet.

The idea is to “organize ourselves into some more specific grand challenges that we can tackle together to have an even greater impact,” said Magdalena Balazinska, director of the Allen School and a UW computer science professor, opening the school’s annual Research Showcase and Open House.

Here are the six grand challenges:

• Anticipate and address security, privacy, and safety issues as tech permeates society.
• Make high-quality cognitive and mental health support available to all.
• Design technology to be accessible at its inception — not as an add-on.
• Design AI in a way that is transparent and equally beneficial to all.
• Build systems that can be trusted to do exactly what we want them to do, every time.
• Create technologies that sustain people and the planet.

Balazinska explained that the list draws on the strengths and interests of its faculty, who now number more than 90, including 74 on the tenure track.

With total enrollment of about 2,900 students, last year the Allen School graduated more than 600 undergrads, 150 master’s students, and 50 Ph.D. students.

The Allen School has grown so large that subfields like systems and NLP (natural language processing) risk becoming isolated “mini departments,” said Shwetak Patel, a University of Washington computer science professor. The Grand Challenges initiative emerged as a bottom-up effort to reconnect these groups around shared, human-centered problems. 

Patel said the initiative also encourages collaborations on campus beyond the computer science school, citing examples like fetal heart rate monitoring with UW Medicine.

A serial entrepreneur and 2011 MacArthur Fellow, Patel recalled that when he joined UW 18 years ago, his applied and entrepreneurial focus was seen as unconventional. Now it’s central to the school’s direction. The grand challenges initiative is “music to my ears,” Patel said.

In tackling these challenges, the Allen School has a unique advantage against many other computer science schools. Eighteen faculty members currently hold what’s known as “concurrent engagements” — formally splitting time between the Allen School and companies and organizations such as Google, Meta, Microsoft, and the Allen Institute for AI (Ai2).

This is a “superpower” for the Allen School, said Patel, who has a concurrent engagement at Google. These arrangements, he explained, give faculty and students access to data, computing resources, and real-world challenges by working directly with companies developing the most advanced AI systems.

“A lot of the problems we’re trying to solve, you cannot solve them just at the university,” Patel said, pointing to examples such as open-source foundation models and AI for mental-health research that depend on large-scale resources unavailable in academia alone.

These roles can also stretch professors thin. “When somebody’s split, there’s only so much mental energy you can put into the university,” Patel said. Many of those faculty members teach just one or two courses a year, requiring the school to rely more on lecturers and teaching faculty.

Still, he said, the benefits outweigh the costs. “I’d rather have 50% of somebody than 0% of somebody, and we’ll make it work,” he said. “That’s been our strategy.”

The Madrona Prize, an annual award presented at the event by the Seattle-based venture capital firm, went to a project called “Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward.” The system makes AI chatbots more personal by giving them a “curiosity reward,” motivating the AI to actively learn about a user’s traits during a conversation to create more personalized interactions.

On the subject of industry collaborations, the lead researcher on the prize-winning project, UW Ph.D. student Yanming Wan, conducted the research while working as an intern at Google DeepMind. (See full list of winners and runners-up below.)

At the evening poster session, graduate students filled the rooms to showcase their latest projects — including new advances in artificial intelligence for speech, language, and accessibility.

DopFone: Doppler-based fetal heart rate monitoring using commodity smartphones

DopFone transforms phones into fetal heart rate monitors. It uses the phone speaker to transmit a continuous sine wave and uses the microphone to record the reflections. It then processes the audio recordings to estimate fetal heart rate. It aims to be an alternative to doppler ultrasounds that require trained staff, which aren’t practical for frequent remote use.

“The major impact would be in the rural, remote and low-resource settings where access to such maternity care is less — also called maternity care deserts,” said Poojita Garg, a second-year PhD student.

CourseSLM: A Chatbot Tool for Supporting Instructors and Classroom Learning

This custom-built chatbot is designed to help students stay focused and build real understanding rather than relying on quick shortcuts. The system uses built-in guardrails to keep learners on task and counter the distractions and over-dependence that can come with general large language models.

Running locally on school devices, the chatbot helps protect student data and ensures access even without Wi-Fi.

“We’re focused on making sure students have access to technology, and know how to use it properly and safely,” said Marquiese Garrett, a sophomore at the UW.

Efficient serving of SpeechLMs with VoxServe

VoxServe makes speech-language models run more efficiently. It uses a standardized abstraction layer and interface that allows many different models to run through a single system. Its key innovation is a custom scheduling algorithm that optimizes performance depending on the use case.

The approach makes speech-based AI systems faster, cheaper, and easier to deploy, paving the way for real-time voice assistants and other next-gen speech applications.

“I thought it would be beneficial if we can provide this sort of open-source system that people can use,” said Keisuke Kamahori, third-year Ph.D. student at the Allen School.

ConvFill: Model collaboration for responsive conversational voice agents

ConvFill is a lightweight conversational model designed to reduce the delay in voice-based large language models. The system responds quickly with short, initial answers, then fills in more detailed information as larger models complete their processing.

By combining small and large models in this way, ConvFill delivers faster responses while conserving tokens and improving efficiency — an important step toward more natural, low-latency conversational AI.

“This is an exciting way to think about how we can combine systems together to get the best of both worlds,” said Zachary Englhardt, a third-year Ph.D. student. “It’s an exciting way to look at problems.”

ConsumerBench: Benchmarking generative AI on end-user devices

Running generative AI locally — on laptops, phones, or other personal hardware — introduces new system-level challenges in fairness, efficiency, and scheduling.

ConsumerBench is a benchmarking framework that tests how well generative AI applications perform on consumer hardware when multiple AI models run at the same time. The open-source tool helps researchers identify bottlenecks and improve performance on consumer devices.

There are a number of benefits to running models locally: “There are privacy purposes — a user can ask for questions related to email or private content, and they can do it efficiently and accurately,” said Yile Gu, a third-year Ph.D. student at the Allen School.

Designing Chatbots for Sensitive Health Contexts: Lessons from Contraceptive Care in Kenyan Pharmacies

A project aimed at improving contraceptive access and guidance for adolescent girls and young women in Kenya by integrating low-fidelity chatbots into healthcare settings. The goal is to understand how chatbots can support private, informed conversations and work effectively within pharmacies.

“The fuel behind this whole project is that my team is really interested in improving health outcomes for vulnerable populations,” said Lisa Orii, a fifth-year Ph.D. student.

See more about the research showcase here. Here’s the list of winning projects.

Madrona Prize Winner: “Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward” Yanming Wan, Jiaxing Wu, Marwa Abdulhai, Lior Shani, Natasha Jaques

Runner up: “VAMOS: A Hierarchical Vision-Language-Action Model for Capability-Modulated and Steerable Navigation” Mateo Guaman Castro, Sidharth Rajagopal, Daniel Gorbatov, Matt Schmittle, Rohan Baijal, Octi Zhang, Rosario Scalise, Sidharth Talia, Emma Romig, Celso de Melo, Byron Boots, Abhishek Gupta

Runner up: “Dynamic 6DOF VR reconstruction from monocular videos” Baback Elmieh, Steve Seitz, Ira-Kemelmacher, Brian Curless

People’s Choice: “MolmoAct” Jason Lee, Jiafei Duan, Haoquan Fang, Yuquan Deng, Shuo Liu, Boyang Li, Bohan Fang, Jieyu Zhang, Yi Ru Wang, Sangho Lee, Winson Han, Wilbert Pumacay, Angelica Wu, Rose Hendrix, Karen Farley, Eli VanderBilt, Ali Farhadi, Dieter Fox, Ranjay Krishna

Editor’s Note: The University of Washington underwrites GeekWire’s coverage of artificial intelligence. Content is under the sole discretion of the GeekWire editorial team. Learn more about underwritten content on GeekWire.

When I was a kid, my mom used to call my Nintendo the “anti-social idiot box.” The widespread assumption back then was that video games, in any format, were a new and particularly efficient way to waste time and money while also becoming an obsessed shut-in.

Over the course of the subsequent decades, video games have grown into both a multi-billion-dollar industry and a much more socially acceptable hobby. While gaming does attract its share of anti-social obsessives, just like any other form of media, I’ve found it’s much more common for people to meet and bond over their mutual enjoyment of the hobby.

Whether it’s friends you meet through MMORPGs or fighting games, finding stories and characters that deeply resonate with you, or discussing your latest game in a shared space like Bluesky or a message board, video games often have a positive impact on the people who play them. That impact simply doesn’t get a fraction of the press of gaming’s various downsides.

That ability is the focus of a new paper from the University of Washington, “’I Would Not Be This Version of Myself Today’: Elaborating on the Effects of Eudaimonic Gaming Experiences.” The paper, by Nisha Devasia, Georgia Kenderova, Julie A. Kientz, Jin Ha Lee, and Michele Newman, was the focus of a presentation this month at the Annual Symposium on Computer-Human Interaction in Play (CHI-PLAY) in Pittsburgh.

For the paper, the authors surveyed 166 respondents about the “meaningful experiences” they’d had as a result of playing video games, such as rich storytelling, becoming interested in specific skill development, or the experience of watching a narrative shift based upon the player’s in-game actions.

According to the paper’s abstract, “While much of the research in digital games has emphasized hedonic experiences, such as flow, enjoyment, and positive affect, recent years have seen increased interest in eudaimonic gaming experiences, typically mixed-affect and associated with personal meaningfulness and growth.”

Of the 166 respondents, 78% reported that they’d had meaningful, life-changing experiences from their time playing video games, the researchers said in a UW News story about the paper.

“We highlighted three conclusions drawn from modeling the data,” Devasia told UW News. “The first is that playing games during stressful times was strongly correlated with positive outcomes for physical and mental health. For example, during COVID, people played games they felt strongly improved their mental health, such as Stardew Valley.”

Devasia also noted that other respondents had developed new interests, such as sports, due to video games they’d played, or gained insight into themselves or their identities from the journeys undertaken by video game protagonists.

“Playing as a character and seeing your choices change the course of events is pretty unique to games, compared with other narrative media like novels or movies,” Devasia said.

“As researchers, we develop games for learning, for instance, for teaching people about misinformation or AI, or promote digital civic engagement, because we want to foster meaningful experiences,” Lee added. “But a lot of the existing research just focuses on the short-term effects of games. This study really helps us understand what actually caused a game to make a difference in someone’s life.”

It sounds obvious at first glance if you’re someone who grew up around video games. It’s almost a given that there’s at least one game that made a serious mark on you somehow, especially if you live in a heavily nerd-coded space like the greater Seattle area.

Anecdotally, that strikes me as an underexplored part of the hobby. If anything, there’s a strange critical drive in the space to deliberately treat gaming as disposable pop culture, without any real meaning or lasting value. If you read any op-ed in the gaming press that discusses the cultural or political meaning of a video game, someone will inevitably show up in the comments to accuse the author of overthinking something that isn’t meant to matter. It’s “just a game.”

Even so, modern video games have just as much ability to resonate with their audience as any novel or film, and people who’ve grown up with them will take lessons away from that. It’s something we don’t discuss often enough in the field; we’ll talk at length about how video games are fun or socially acceptable now or a surprisingly big business, but their influence as culture is less discussed.

“People have a tendency to treat technology as a monolith, as if video games are either good or bad, but there’s so much more nuance,” Kientz told UW News. “The design matters. This study hopefully helps us untangle the positive elements. Certainly, there are bad elements — toxicity and addictiveness, for example. But we also see opportunities for growth and connection.”