Where are the field builders?
Appreciating the unsung heroes of technoscience.
Over the past few years, we’ve been running science funding experiments on Experiment—supporting lots of scientists and technologists with small grants. We’ve gathered data, heard stories, and answered many of the initial questions we posed at the outset:
Do small grants matter? Yes, definitely.
Would the “science angel” model work to find overlooked people and ideas? Yes, and more could be done.
Could grant funding spur more crowdfunding? Not in a meaningful way, but on occasion, and maybe we could do it differently.
Does seed funding lead to follow-on funding from larger science funders? Yes—it’s still early but it’s happening.
I could write a long essay about each of those answers. But the more interesting observations came from questions we never thought to ask. At this point, those ideas are closer to strong hunches rather than testable hypotheses, but they are the truest lessons we have from our experience. Here are two of them.
The first idea comes from working with more than two dozen science angels—scientists-turned-funders who have the discretion to allocate small grants to projects they deem worthy. I was constantly surprised by the varying approaches of these angels. Some struggled to get grants out, while others easily accepted the challenge. I have an emerging intuition about the characteristics of a good science funder (which I’ll explore below), but I’m stuck on this thought: we still can’t predict who will be good at this.
We aren’t the only ones who are unsure; nobody knows. Being a good science funder seems to be a completely separate skill from being a good scientist. And outside of ARPA program managers, there is very little effort to identify or develop the talent.
The second idea isn’t from our work specifically, but our type of work more broadly. As we’ve been operating in the uncharted area between academia, tech startups, and government agencies, we’ve found fellow travelers—other individuals and groups operating in the same in-between space. Many of these people identify as scientific “field builders” and they specialize in generating momentum among funders, technologists, and researchers toward some new field of technology or science. They drive funding, make connections, and facilitate progress toward real-world outcomes—not just published papers. Companies, tools, and industries are birthed in their wake. Field builders are opening the doors to new possibilities.
Seeing these operators up close—mostly through the Experiment Foundation’s lens as a peer organization—has profoundly affected my perspective. I’m convinced that technoscientific progress is being hindered by a lack of field builders.
This essay explores these two ideas—their origins and possible explanations. It also sets up the big question we’re considering at the Experiment Foundation, emerging at the intersection: can we use the science angel model to embolden more field builders?
“OK, you do it.”
Scientists love to gripe about money. In both private circles and public forums, scientists jump at the opportunity to explain their vision for how federal budgets should be managed, how quickly grants could be administered, or how new funding mechanisms be employed. Their opinions seem to grow stronger with age and experience.
"Money problems" topped the list of a survey of academic researchers in 2016. I suspect you'd get a similar result if you ran the questionnaire again today. I spent a year interviewing scientists and mostly heard the same. The final question of my interview was always, "What is your best idea to make science better?"
And the answer was almost always a variation on more diverse funding models.
Our science angel funding program has challenged scientists to become the science funder of their dreams by giving them discretion (within the bounds of scientific legibility and non-profit legality) over a small budget, the ability to make quick decisions and a simple charge: OK, you do it.
Here is the original pitch:
Foundations and companies seed fund the program. A group of scientists are selected and each given a budget of $50-100k to contribute to projects on Experiment. They are free to use their discretion in how they recruit, select, and allocate that amount. At the end of a one year period, they will have a portfolio of experiments to show their work. All of this is done in public on Experiment, with minimal overhead and the potential to leverage additional support from the crowd.
Three years later, I'm proud to say we ran the experiment. Bold philanthropic partners helped us enable more than twenty individuals (and occasionally small teams) to participate as science angels and we've funded nearly 250 projects as a result of the program. The initiative has even won awards: our partnership with Robert Downey Jr's FootPrint Coalition won the 2022 Falling Walls "Breakthrough of the Year" in the Scientific Management category.
After three years of running these programs, my perspective on the problem has changed. I've lost interest in trying new and exotic funding mechanisms, like lotteries. Some of these will work, I'm sure, but my forays into these ideas have been underwhelming. Instead, I'm convinced the big opportunity for fixing science is by improving funding dynamics, especially at the earliest stages.
The best part of the science angel program was giving scientists the freedom to experiment with those dynamics—the subtle human aspects that happen before, during, and after the actual grant funding. Their efforts were revealing.
The characteristics of a truly great science angel are still up for grabs, yet to be discovered. The metascience of the question is worth further study. But we’ve found a few basic truths—the ABCs—to screen for the right types of people.
A - Availability. This seems obvious, but it's an easy mistake. To be a good science funder, you need to dedicate time to the endeavor. A lot of smart people take on too many commitments, and overestimate what they can accomplish. In the face of overwhelm, a part-time science funding gig is one of the first candidates for procrastination. Famous and established academics sometimes turn out to be ineffective science angels, mostly because of their availability, or lack thereof. They're too busy.
We could improve here. We could create an interface and cadence that works better for the busy schedules of established scientists, but I’d rather focus effort on those who can dedicate time and energy to the job. Besides, I’ve learned there are more important characteristics to optimize for, like…
B - Belief. Done well, science angel-ing is more than money. Like their startup-investing counterparts, they also inject a shot of enthusiasm and momentum into a promising (although sometimes rough) idea. The best science angels can imbue a sense of possibility alongside the grant. These are the grants that change careers. Here's Nobel Prize-winner Katalin Kariko writing about her first true believer, David Langer:
“What was unusual, though, was how deeply David believed in the work. Not only had he managed to secure a small grant—about $25,000, if I recall correctly—for our work, he’s also delivered a paper at a conference in Arizona: “Bypassing the Nucleus: mRNA as Gene Therapy.” I guess it’s fair to say that David wasn’t merely a believer in mRNA—he’d become an mRNA evangelist.”
Money can be a vessel for belief, which is often more sustaining. Conveying belief—authentically filling another with self-confidence in their nascent idea—seems to be an innate talent, almost a personality quirk. Startup investors with this skill tend to rise quickly. Unfortunately, the behavior is discouraged in the scientific ranks. Scientists are trained from an early age to be critical and skeptical of their peers. In science, belief is never rewarded and sometimes punished. But it's essential to being a good angel investor, whether financial or scientific.
Rigor is good, welcome. But early-stage ideas, especially when they're novel, need a heavy dose of belief when they begin their journey through the gauntlet of criticism and peer review.
My best idea to identify this trait is to go earlier—start more folks as science angels early in their careers and see if the job suits them. If they have the knack, we should provide an opportunity for career progression, with the best rising into the talent funnel at one of the emerging ARPAs.
C - Curiosity. Like availability, curiosity is an obvious characteristic, but you'd be surprised. Unfortunately, many scientists become so thoroughly absorbed in their milieu of papers, grant paperwork, and lab management that they lose the playful curiosity of the beginner's mind. They become hyper-fixated on the cutting-edge novelty of their specific academic focus—the new papers, competitive research, applications of AI to their field, etc—and lose the ability to see possibilities outside their current perspective. An effective science angel must have real curiosity towards new ideas or research directions, even and especially if they're not fully fleshed out. They should point out and nurture the surprising bits of a proposed question, rather than getting hung up on potential pitfalls. This characteristic creates a feeling of co-conspiring with grantees, as opposed to pure critique.
I don't know how to test for curiosity yet, but it's worth creating some sort of filter. So far, the best wisdom here comes from a short story from Holly Witteman: avoid B Lane swimmers.
And a new addition: D - Different. I recently asked Bridget Baumgartner, an experienced trainer and coach of ARPA program managers, what she looks for in a potential funder. She replied with a simple heuristic: they must have a vision for their field or the world that is fundamentally different from what exists today.
“How will the world be different if you’re right?” is a simple and potent question to ask a potential science angel. We incorporated Bridget’s insight into our screening process immediately.
These lessons—the ABCs—are just a baseline filter. They help us identify who won’t succeed in the role. But there’s still an upper bound to explore: who can be great. To answer that question, we need an ideal—a benchmark or target that we’re aiming to emulate or exceed—otherwise we’re just wandering around in the dark. Enter the field builder.
The Technoscientific Field Builder
We aren’t alone. Our science funding experiment is one tiny star in a constellation of new organizational and institutional experiments playing out in research.
The broad strokes of this movement and trend are clear. A frustration with publication-obsessed academia and a growing opportunity in startup venture creation has pushed talented scientists into the driver's seat. No longer wanting to be stuck in the lab, endlessly publishing papers, their scientific dreams have turned toward building companies, low-bureaucracy research autonomy, and impact. The best historical analogy is 1950s Hollywood—the moment when the studio system broke down and a new, talent-centric model emerged.
The best scientists have options. They're moving between non-profits, universities, and startup circles—often all at once—to build their purpose-fit empires. And these entrepreneurial scientists are paving the career path just as fast as they are carving it—with help. Buoying many of these efforts, working alongside the entrepreneurial scientists, is an emerging persona: the field builder.
The field builder operates between the insular worlds of academia, startups, and non-profit philanthropy. Sometimes they are giving grants. Other times they are educating investors. But they're always relentless cheerleaders for a new and emerging field of technology or science—their enthusiasm acting as a siren call to other researchers, builders, and doers to get involved. To invoke the Hollywood analogy again, the field builders are serving the same enabling role that talent agents played in show business. Agents helped advocate for talented actors, directors, and writers—they facilitated a grand renegotiation of power and process.
Crucially, in the aftermath of Hollywood's revolution, from the Golden Age to New Hollywood, there was a renaissance in filmmaking. We should hope for the same in science. And field builders are making it happen.
I'm thinking of Isha Datar, who coined and kickstarted the field of cellular agriculture. Or Ryan Phelan, whose entrepreneurial effort brought biotechnology into the world of wildlife conservation. Or Antonius Gagern who helped bring ocean alkalinity enhancement into the forefront of carbon removal discussions.
If done well, field building can be wildly generative. Once they start the flywheel, the usual suspects show up to go bigger: philanthropists and researchers, entrepreneurs and investors, governments and policy-makers. Datar, Phelan, and Gagern have all helped catalyze communities of practice and directed millions of dollars into their respective fields. Many of those initial investments have been eclipsed by entrepreneurs, funders, and government agencies that have taken those nascent ideas to larger scales. Field builders are fire starters.
The term “field building” is gaining traction in philanthropic circles. The Bridgespan Group, an advisory firm that monitors and advises many large foundations, has been promoting the concept for decades. Their most recent report was published in 2020, Field Building for Population-Level Change, which describes and promotes the concept as a vital philanthropic strategy. They were using it in the context of social change, applying it to causes like democracy, bail reform, or homelessness alleviation. Their 40-page report, which analyzed the development of more than 36 different fields, reads as a good introduction to the idea. There are some useful contributions, like segmenting field building into distinct phases, but the tactical recommendations seem vague and tentative. It’s still early days.
The term has a somewhat separate meaning and significance in technoscience, and it’s newer. More than a philanthropic strategy, it’s a job to be done—a persona to inhabit. The stark boundaries between academia and the marketplace, non-profit and for-profit, have opened up a grey area where field builders can make a large difference.
The momentum of the field building concept in science owes a debt to Tom Kalil’s definition and promotion. Kalil has spent his career in the upper echelons of the ivory tower as an administrator at the University of California, Berkeley, director at the White House Office of Science and Technology Policy, and most recently as an executive at prominent tech-focused philanthropies. I first heard the term from him—he calls them field strategists. Throughout his career, he noticed the rare person who can take a longer, wider view of their discipline, whether by identifying common bottlenecks or opportunities or simply seeing a new direction a field could move in. Recognizing their outsized impact, he bet big on new organizational structures like Focused Research Organizations to support these outliers. His encouragement has also helped to create new resources on the topic, like Ed Boyden and Adam Marblestone’s paper, “Architecting Discovery”, or Eric Gilliam’s essay, “When do ideas get easier to find?”
Gilliam’s message, in particular, speaks to the value of field builders. He documents the generative nature of the work, but also highlights the paradox of the current system: the incentives in science go against creating new branches of knowledge in favor of digging ever deeper silos of understanding.
There’s more to learn, too. A good place to start is by asking and studying the people who are doing the work. I’ve started a crowd-sourced list of technoscientific field builders to help with that process. There’s an emerging suite of tactics being employed. The field builders are in contact with each other now, recognizing their shared philosophy and overlapping interests. For example, here’s Homeworld Collective writing about their field building methodology, and the tools they’ve utilized to be effective.
I’ve talked to many of them and the commonalities go far beyond strategy. They’re all onto something. Listening to them describe their mission, one gets the sense they are sprinting to catch up with the opportunity. And they’re eager to share their playbook so others can get involved.
Building Field Builders
The backgrounds of these field builders might surprise you. More than half of the people listed didn’t come from prominent labs or research institutions—some have no formal scientific training at all. And many who did have a scientific education also had important life experiences outside of research. Datar credits her party planning experience. Phelan was a biotech executive. Gagern worked as a consultant after studying economics. Unsuspecting people are thriving in the role.
This matches my experience working with science angels. It’s hard to correlate their effectiveness with any sort of traditional measure of scientific achievement or advancement. The trend suggests we aren’t casting a wide enough net. And given the efficacy of technoscientific field building, more serious questions arise: how do we get more people to do it? And from where do we draw?
Before getting to specifics, it’s worth reiterating that field building is a positive sum game. In nearly all the examples above, the field builders made something from nothing—they brought in new funders, inspired new entrepreneurs, and gave opportunities to more researchers and technologists. Scientists are so accustomed to fighting for their slice of resources, like grant funding or talented grad students, that they often react skeptically to activity that grows the entire pie.
In that spirit, there may be many ideas for creating more field builders. I’d like to hear them. I also want to pitch our idea: turning our small science angel experiment into a larger, full-on field building accelerator.
Almost all the great field builders get into grant funding. If they don’t start as funders, they tend to evolve the trait quickly. Distributing money and resources to a growing scene seems to be a core competency. Running a micro-grant program, a la science angel-ing, is a good way to isolate and evaluate the ability for larger amounts—it’s a perfectly sufficient starter pack.
Our model is particularly well-suited to the job. Here’s what we could do:
Philanthropic partners underwrite a cohort of 50 aspiring field builders. Each gets a budget of $75,000 to start micro-granting out to new ideas and researchers in their respective fields. We run a bi-weekly interview series with experienced field builders to tell stories and answer questions. At the end of one year, we’ll assess how their efforts impacted the field.
We have the administrative and legal infrastructure to run this experiment tomorrow. Added twists and details could make it even more fun, interesting, and impactful.
In the final essay of this series (coming next), I’ll explore this pitch in more detail—how we’d structure it, how we’d measure it, and what aspects still need consideration.
If you missed the first installment, here’s part 1:
Small, Fast Grants at the National Science Foundation
feels like a warren weaver thing...https://www.freaktakes.com/p/a-report-on-scientific-branch-creation
The Climate Party - https://www.linkedin.com/company/the-climate-party/ - in India is driving a cross-disciplinary approach to building crucial climate solutions by making unlikely people (already working on different problems) to collaborate.