Steve Brotman: Hi, I am Steve Brotman, the Managing Partner at Alpha Partners. Welcome to Driving Alpha, where we explore the strategies and stories behind the world’s most influential venture capitalists. Today, we’re joined by Francisco Gimenez, a partner at 8VC, a firm founded by Joe Lonsdale after his success with Palantir. I think he started, what, five or six unicorns? And it’s dedicated to using entrepreneurship to solve the world’s most difficult problems. Quite a thesis there. Francisco’s journey is a testament to the power of interdisciplinary expertise. He earned his BS in Electrical Engineering and Computer Sciences at UC Berkeley and went on to get a PhD in Biomedical Informatics at Stanford. Before joining, he served as Resident Data Scientist at Formation 8, working hands-on with portfolio companies to strategize, prototype, and recruit for data-driven products. At 8VC, Francisco’s been instrumental in shaping the firm’s vision in the biosciences arena. Among some of those companies are BigHat, Unlearn.AI, and Seleno, and they incubated Cambrian and Lattice. And I’m super excited to dive into some of those companies. Francisco is recognized not only for his technical abilities but also for his ability to spot and nurture the next wave of industry-transforming startups. Get ready for an insightful conversation with Francisco Gimenez of 8VC right here on Driving Alpha. So, welcome, Francisco.
Francisco Gimenez: Thanks, Steve. It’s lovely to be here. I don’t do an enormous amount of podcasts, but I really enjoyed all the conversations we’ve had over the years. It’s always been really fun. So happy to join here.
Steve Brotman: Excellent. You know, partially, this podcast is inspired by those conversations, so it’s great to have you, and thanks for joining us. So, one fun thing I like to start off with is I went through your bio, but I’d love to hear how your entrepreneurial journey led you to Formation 8 and then 8VC. You know, people don’t necessarily wake up and say, “Oh, I got my PhD. Let’s go into venture capital.”
Francisco Gimenez: Yeah.
Steve Brotman: Usually you’re kind of focused on a whole bunch of stuff. So, I’d love to hear more about that.
Francisco Gimenez: It was never the plan. I’m not a career planner. My wife is a writer and so there are two kinds of writers. They’re called planners and pantsers. Planners will outline the whole story and then pantsers just write by the seat of their pants. And so I’m a pantser as it pertains to my career so far. You know, I started in computer science just because I loved computer science. And that was in 2005 when it was, kind of hard to believe, but it was considered a terrible field to enter because of the dot-com crash and outsourcing scares and everything. And I think it was the lowest enrollment that Berkeley had over the years, or at least in that dip. But I fell in love with the idea of AI then, which was also considered a scam field. And I think the one thing I had going for me was I was kind of like a hipster about cool technology. And so, I went there, honestly. I loved the intellectual knowledge, but I didn’t do particularly well because I just didn’t have a mission in my work. And so I took some time off at Cal. I actually went to UCSF and that’s when I worked in a gene therapy lab for Parkinson’s disease. And that was a really formative window. It was funded by Andy Grove when he was writing all those articles about how biology is stupid. If they just worked like engineers, they would be so much better. And so they hired a computer scientist like me just to satisfy whatever requirements he had. I think biology is a very different field than engineering and computer science, but it was fantastic to get that opportunity to work with that group. And so, I fell in love with it and I fell in love with image analysis and AI in image analysis. And so, I went to Stanford in 2010 to do that in the Biomedical Informatics program. That was also a backwater field. Anybody who was in image analysis in 2010 knew that it was basically just duct tape and bubble gum, and nothing ever actually worked in production. And I was again lucky enough to be saved by AlexNet in 2012, which basically just revolutionized the field. And to be at Stanford when that was happening was really special. I think we were one of the first, say, medical imaging groups to get a GPU from NVIDIA when we just emailed them and said, “Hey, can we have one?” One of my protégés there wrote one of the earliest papers using AlexNet and deep learning in mammography. And that was really, really cool. And so, I continued to fall in love with that and work hard. And just midway through, I did the same thing. I said, “I want to see a little bit of the rest of the world.” And so, I didn’t know how to get a job or an internship or anything. And I had a friend who said, “Let me just introduce you to a couple of folks.” I think that’s kind of the benefit of going to Stanford. And I got hooked up with Formation 8 at the time, which was Joe’s predecessor fund when he was doing a lot in data science and data engineering in his smart enterprise thesis, which we call vertical SaaS now. And it was supposed to be just a summer internship, really. It was just that I liked the idea of learning about entrepreneurship because you just see it from a distance, but it was hard to intuit it. And I really just loved it. I got lucky enough to work on some interesting projects that were successful, and then they asked me to stick around part-time while I finished my PhD. And when 8VC got started, they asked me to join the team as a data scientist and then quickly just moved me on to the investment team. And I’ve been here ever since. I think just following whatever I kind of think is cool has been helpful for me, at least.
Steve Brotman: Wow. Well, with the starting point, solving the world’s most difficult problems, that’s quite the gauntlet. And just noting, you know, what you’re talking about—AI 15 years ago, image scanning, all that stuff, GPUs—these are all things that were way out there on the fringe. Even, you know, four or five years ago, they were pretty fringy. What’s interesting at Alpha is we’ve done, I think, three or four health tech deals, which you and I talked about, just because the vortex of opportunity is just so damn big in the life sciences arena to apply technology to everything—from just in terms of process and applying knowledge that we already know. It sounds like you’re very much a visionary early on in your career, way before anyone else really was focused on this. What’s exciting you the most in this whole arena that we’re a part of? You mentioned some of these companies and you’ve incubated some companies. What’s getting you the most amount of excitement lately?
Francisco Gimenez: I’ll have one quibble. I don’t know if I was ever a visionary. I just worked on things that were really, really interesting to me personally. And, look, I’ve worked on plenty of things that were interesting to me and didn’t manifest as anything, so it’s just—but if anything, I probably lacked a little bit of the vision. I should have bought a bunch of NVIDIA stock in 2012.
Steve Brotman: You’d have been retired by now, right?
Francisco Gimenez: Yeah, exactly. Should have pushed us to invest in OpenAI when I saw the first—you know, we were one of the first groups to get access to GPT-3. It was the summer of 2020. And then there was the Slack group where Brockman would give you access to things. And I think there’s a lot that I love and see and love to play with, but I think it’s great to be a visionary VC. There are some great ones out there. I think, though, it’s better to be a VC who recognizes visionary founders and has them teach us about the future because they’re the ones who have to do this sort of black magic of inventing it. So, I think that’s hopefully where I’m leaning into, as opposed to trying to be the thought boy.
Steve Brotman: I think that’s a really good point. Are there any things that pop out as you’re talking to entrepreneurs? Like, is it being more of an artist and looking up on the wall and saying, “Wow, that guy’s super visionary?”
Francisco Gimenez: So, you know, I think this is my personal perspective, and I’ll—I mean, I’ll answer your original question, what’s exciting me. But I think for me personally, I would say the companies I love to invest in are ones where they’re tackling a really important problem with a large market that’s really hard, just really, really hard. And if they succeed, they would be an “n of one” company and really unique and only in class. Now, that sounds pretty generic, right? Everybody would say that. But I think the counterpoint there is, you know, I’m maybe less inclined to invest in companies where there are several and I have to pick the best one. And I have partners here who are really fantastic at that, right? So, I’m really glad we have different sorts of investment styles. But, you know, I think—maybe I was seeing a Dean tweet the other day about how Ramp was started when Brex had already raised 300 million. I don’t know if that would’ve been a deal that I would’ve recognized.
Steve Brotman: No, we missed it personally, just so you know. It’s sort of on our—definitely on our anti-portfolio list. But, you know, typically in tech, it’s often the case that the category leader stays the category leader. It’s rare that they’re surpassed. I mean, obviously Google is the 15th search engine to enter that sector, right? So it is possible. But your focus is more one of one.
Francisco Gimenez: Yes. It’s so insane or so creative that nobody would even think to start this company. And I think that’s what gets me really excited. I think Unlearn is kind of the canonical example of that, at least it’s further along that I can speak about, where, you know, the idea of using digital twins to simulate what a patient would look like in a placebo arm of a trial so that we can eliminate the placebo arm of a trial—you know, it’s just nobody was thinking about doing that. There were variations of how people were thinking about reducing trial recruitment, but just really just simulating—putting a patient on a cancer or, say, Alzheimer’s drug and simulating what they would have looked like. So we only have to, you know, recruit half of the trial, and we don’t have to put anybody on a placebo. I remember when I was doing diligence calls with pharmas, they were just like, “There’s no way this will work. This is insane.” And then my follow-up question was like, “But what if it did work and it was approved?” And they’d be like, “Well, I’d be fired for not using it.” And I think that was like the perfect encapsulation of a deal that I love to do. There’s no way it’ll work, but if it does, I have to use it. And that’s kind of what I think about as an “n of one” company and tackling a really difficult problem or kind of an obscure problem. I think a lot of people come to me, and I get really excited when people teach me about, “Hey, there’s this massive area of spending,” or, “There’s this massive area of biology that’s unrecognized or unloved.” But if we do it, it’s pretty transformative.
Steve Brotman: Interesting. So, it’s mostly more thesis-driven, more like just the direction and the vision of that founder into what they perceive might be a whole new world?
Francisco Gimenez: Yeah, I mean stage-wise, yeah. Right? Because, right, if it’s in the early stage, it’s a great founder, there’s a really interesting problem, and there’s kind of an existence proof or theoretical solution that we can agree would possibly tackle that problem, that’s maybe one way to view it. At least if I’m, like, looking back at the things that I’ve done that seem to have been working, that’s what manifested. Starting with, for example, interesting technology maybe finds a solution, but frequently it doesn’t. And so I think it’s really—if you start with a problem, then, and you keep that as kind of the guiding principle of the company, more often than not, we find a way through. And I think that’s honestly something I’ve learned from Joe. If you really just sit down with him, Joe thinks about the world in terms of problems to solve and that companies are the vehicles to solve those problems.
Steve Brotman: Interesting. You mentioned great entrepreneurs, and I mean, I think that a great entrepreneur with some of these—I hate to use the word —kooky”—but end-of-one things, it’s sometimes hard to distinguish, right? And especially, I mean, it certainly helps that you got your own PhD in that and have an undergrad in computer science. So, like, that sort of helps you navigate that world because there are some folks that just want to research. They just want to, you know, that’s all they really want to do. What’s your thought on those folks forming companies? Like, do you need to pair them with somebody else? Is it usually a group? Is it a team? Is it a single founder? So, I’m just sort of curious because I’ve wandered the halls of some of these universities, and I’m not a bioscience investor, but even just trying early in my career, I just felt like I just couldn’t find those people. I couldn’t find my entrepreneurial people. Maybe it’s different out at Stanford on the West Coast, where they just—that’s just native—but at SUNY Albany, they’re pretty happy being a doctoral student or going on to get, you know, their postdoc and continue to do research. Yeah. So, I was just sort of curious, how does the entrepreneur typically manifest? Is it a serial entrepreneur? Is it the first time? Is there no pattern? That’s okay too.
Francisco Gimenez: Look, if we did this interview five years ago, I probably would’ve had a more prescriptive answer. And I think this job has actually dissolved me of theoretical answers here. And so, I think the through line—and I’ll borrow a phrase from Rick Klausner, kind of a luminary in oncology who was the head of the NCI and is the co-founder of Altos, which is a longevity company we’re invested in—and at least his criticism of the biotech space, and I think this rings well more generally if we think about deep tech or difficult technologies, is that everybody thinks in terms of projects and not problems. Or at least the majority do. And so, in biotech, you know, you think about the project of, here’s this maybe indication, but here’s this molecule, and we just have to figure out our checklist to de-risk it in a stage-wise manner. And it’s really about not making mistakes and executing the plan pre-specified. And then, you know, it’s kind of like rolling the dice and seeing where they land, you know, in the sense of a clinical trial or something. And I think the founders who I have been drawn to are those problem-oriented founders. They fall in love with the problem because there may be many ways to attack it. And if you fall in love with the project, you’re stuck on a certain way to solve a problem. But if you’re attached to the problem, you’re going to find any way you can to solve the problem.
Maybe the best sort of material in this, or maybe where I’m probably paraphrasing it, is Paul Graham’s essay last year, The Right Kind of Persistence, where he talked about the difference between persistent and obstinate. And he summarizes it as obstinate as being attached to the way you’re solving a problem and persistent as being attached to solving the problem. And that small, almost seemingly semantic difference is, I think, the huge difference that I’ve seen manifest. I think that was probably one of the most interesting insights that I’d read by him. He wrote the Founder Mode essay immediately afterward, and that got all the attention. But I think The Right Kind of Persistence essay was possibly one of the most interesting things I think he’s written in a long time, from my perspective.
Steve Brotman: That’s pretty interesting. And a lot of—I mean, and I don’t mean to broad brush, but a lot of PhDs, how many of them are equipped with that problem-solving mentality and be willing to leave the womb of academia, right? Do you find that to be a problem or not so much?
Francisco Gimenez: I don’t think it’s too much of a problem about enough PhDs, you know, leaving. I think that science teaches you a very specific way of thinking that can have a tension with entrepreneurship. And if you’re starting a science-backed company, obviously you need to be able to think like a scientist. So that’s not to say that you are prevented from it. It’s just in science, I think the core perspective is that we have to consistently doubt our findings and we have to consistently check it for truth. And we really have to hold a healthy skepticism in what we’re doing. And I think in entrepreneurship, you have to be incredibly optimistic about your ability to solve a problem. And if science is taught poorly or not taught with a little bit of that notion of optimism in mind, it can kind of drive you into this perspective of like, nothing ever works, nothing ever happens, everything has a caveat and whatnot. Versus just saying, “I can solve this problem. There’s something meaningful here.” And I have a reason to believe it doesn’t invalidate you, but it’s just a different way of thinking that can be in tension.
Steve Brotman: You know, having an engineering background, which is more problem-focused versus sort of theory-centric, like trying to find new knowledge—have you noticed a pattern there where if there was an engineering training, that was helpful?
Francisco Gimenez: Well, we certainly have the tools and mechanisms and sort of talk track for how to teach engineers how to think about entrepreneurship. I think that’s what we’ve been doing for, you know, as an industry for some decades now. And I think that could be done a little better with respect to scientists because, you know, there are plenty of scientists entering entrepreneurship, but again, it’s very project-focused. If you have your kind of prototypical, say, life science VC incubated company, it’s about not making mistakes and following a plan as opposed to this like pivot and iterate and try to figure out how to solve this problem, right? And there are structural reasons why you can’t pivot a company three times when you have long cycle times for findings. But there’s a certain amount of that thinking that I think is not innately trained or inherently trained.
And so, whereas engineering—like, let’s say, computer science—there were decades spent in sort of the waterfall model of programming where everybody coded their specific piece and then we, like, mushed it all together and hoped it would work. And that was, you know, repeatedly shown to be a total failure of how to do anything. And this sort of emergence of agile programming was like, let’s build the skinniest end-to-end functional version of a program and then add and layer onto that because we’ve just assumed a certain amount of inefficiency and incorrectness. And the only way to solve for that is iteration. And in science, you know, you might say, like, here’s our big hypothesis, here’s some biomolecular target that we’re going after that might have an implication in cancer or something like that. And you just have a longer inherent time to validate that target, to make a molecule, to test it in some kind of different models with increasing fidelity of predictability in humans. So, iteration is a little harder. But I think we get so backed up in that where we have this almost waterfall style of bio-therapeutic development, where, you know, a company might run up into their sort of big bang experiment of in vivo models and hope it works and have just enough runway to prove that out as opposed to what we would like to do, which is just frequent, rapid iteration of testing that allows you to de-risk an experiment or a thought process.
Steve Brotman: So, going back to what we were talking about earlier, which is what most excites you today, is there a key— you know, a couple companies you might want to talk about or—
Francisco Gimenez: I mean, companies—the ones that you’ve mentioned are ones that are pretty exciting right now. So, BigHat Biosciences is sort of engineering biologics using traditional wet lab approaches and AI approaches for all manner of therapeutic development. And they were, you know, launched as kind of an idea on the back of a napkin in the summer of 2019. And now AI in drug discovery and AI in protein design at large is a very sort of hype-y field, and the foundation models for biology companies and all that other stuff. But I think what they’ve built is really differentiated and exciting in that space. I think by, you know, the number of companies they’re partnering with—both publicly stated and unrevealed so far—and then their internal programs, they were early to what I think is the right way to engineer biologics.
And in particular, you know, one of the co-founders, Mark, told me very early on when I was in a board meeting, I was like, “How many, say, antibody constructs can we test in a cycle time?” You know, be it like a month or a month—and I was like, “Is it like thousands? Hundreds of thousands? Millions? Billions?” And he said, “This is the wrong question. The right question is, what is the latency of our experimental cycle time? Because I would rather have 200 great experiments in two weeks than a billion weak experiments in two months. And if you truly trust your AI, then those 200 candidates you’re putting into experimentation are really compelling. And if you truly trust your active learning framework or feedback loops, you want to get that readout quickly so that you can optimize over them much more quickly.” And so again, it’s kind of like this notion of iterative development. They found a way to do it with their platform that I think has really borne out in the accomplishments they’ve done, both with their partners and internally. And so again, it’s not just bringing computation into the life sciences. Everybody thinks of that when they chat with me. They’re like, “What about AI, drug discovery, blah, blah, blah.” But I really think it’s about bringing engineering principles—the good parts of engineering principles we’ve learned in software companies—into the life sciences. I think that is actually the core component and an important piece. So, I think BigHat is kind of a canonical example of that in the therapeutics landscape.
Seleno, I think, finally had a minimal amount of press. They’ve been in stealth for a few years, but it’s basically all the sort of greats who came out of Illumina building out novel tool systems, and they’re building out a really cool box that allows you to do cell biology manipulation and characterization at kind of an unprecedented depth. You know, allowing you to look at a single cell and run multiple assays on that individual cell and then expand that to cell-cell interactions, groups of cells, and everything. And just this promise of, what if—you know, I like to joke, like, what if cells were the size of soccer balls and we were actually able to manipulate them and move them around like we do with engineering? Maybe a very reductive perspective of mine is like, look, if cells were just larger, we’d be able to do a lot of science a lot more easily. A lot of our rate limiters—it’s just the fact that even just to see what’s going on, we have to do an enormous amount of work, and even just to manipulate them in a certain fashion, we have to do an enormous amount of work. And they just give us those tools to almost be able to manipulate cells with high fidelity and see them and run multiple assays on them.
And I think that as we see more and more tooling in biological foundation models and a lot of the AI systems that are really starting to bear early fruit here, we quickly will move into a regime where we have to have more fine-tuned control of the data we generate for these systems. And I think the analogy I give—though there’s kind of debate about this—is Tesla versus Waymo. Waymo got to kind of full autonomous self-driving first because they had multimodal signals; they had lidar, they had cameras, they had various different things. And they controlled the mechanism for data collection by virtue of having the drivers come in and kind of fill that in. And Tesla was kind of taking what I think is the long-term correct approach, which is let’s put a bunch of basic cameras on cars and just scoop in that data from a bunch of, you know, our customer drivers. And I think Waymo got to fully autonomous first because they were able to have fine-grained control over that.
Now, there are debates about whether in the long run Tesla will win or not. But I think today, at the stage we are with biological foundation models and AI and biology, we need a much more fine-tuned, multimodal view into cell biology to really train and refine and honestly evaluate these models better. And I think Seleno is the platform that’s going to allow us to do that. So, really, really excited about that.
Steve Brotman: Actually, it just makes me think a little bit. Do you feel like your companies are more on the engineering side? Are you doing some basic research? I mean, I’m not a bioscience investor, but what you just talked about seems like you’re doing some pretty basic research or providing tools that other basic researchers can use, which is typically the domain of like the NIH, right, and government.
Francisco Gimenez: So, yes and no. I mean, look, I go to a larger portfolio of things that are more software, you know. Maybe kind of our core thesis for our life sciences group is twofold. One is what we call products, and the other we call infrastructure. And products being the sort of end result of life science products—so biotech, diagnostics, and devices—in that order. And infrastructure is the enabling technology that empowers the bioeconomy to be faster, more efficient, and more productive. And I think that frame—that problem statement for us—it’s, how can we enable more interesting basic science research to translate into people more quickly and more cheaply?
So, I think that the current state of biotech is a very rational manifestation of a system, but it’s not, I think, the best way that we could be. Biotech today is predominantly incubated companies by VC firms that are made to get to sort of a clinical proof of concept that then sell into pharma to fit into and enrich pharma’s constantly at-risk revenue models. Because when drugs get off-patent, they lose essentially that revenue to genericization, which is great. That’s the contract that biotech and life sciences have with society. Arguably, the only deflationary part of healthcare is the fact that drugs go off-patent and get genericized and drop cost over that 20-year time horizon.
So, what does that mean for pharma, though? They have all these retained earnings. They have an incredibly low cost of capital from these drugs that they’re selling that are on-brand, and they buy biotech companies to replenish their pipeline of products. And so life science VCs are very rational, and they think about, you know, if it takes 15 to 20 years from an idea—from conceptualization into drug approval and commercialization—that’s longer than a fund time horizon. And so funds can never build biotech companies off of a promise of cash flows. The system doesn’t allow for that.
And so instead, they figured out a really thoughtful way, which is, let’s get things into clinical proof of concept. And pharma buys these programs when they see evidence. And then pharma really takes those through the really big, meaty phase three trials and then commercializes it, which is really the engines that they have that are sort of unparalleled.
The best sort of analogy I can bring to tech folks is Cisco in the Chambers era doing 150 acquisitions to build out all of their product profiles. You know, it’s like ex-Cisco employees would leave, start a company, and sell it back to Cisco. That’s actually very similar to what happens with pharma. Or even, you know, the Instagram acquisition—I think when Systrom left, the strategic blog post, I think it was Stratechery, Ben Thompson’s blog, said that they were—Instagram was really good at building the product, but Zuck was really good at building the business of Instagram. And I think that’s the right way to think about biotech and pharma—is that biotech’s really good at building the product and pharma’s really good about building the business on that molecule.
Now, that said, that manifests as something that I think is a detriment to society, which is if we have 20 to 25 really truly acquiring pharmas, then we only have 20 to 25 CSOs—chief scientific officers—really determining which diseases we’re going to cure over the next 10 years. And it’s got to be on that pharma time horizon and talk track. So we end up with a lot of biotechs that go after the same targets that are considered validated on topic right now. There are maybe a trillion obesity companies today, and then there are a trillion ADC companies. And target crowding is massive—like, I think somebody was giving me a statistic about how 80% of pharmas have overlapping targets that they’re going after.
And there’s a big debate in the sort of life sciences community as to whether that’s good or bad. And I’m kind of more in the “that’s not very good” camp. But the right way to solve that is we have to push innovation down and decentralize it, which means we cannot have the barrier to drug innovation be pharma decision-making. It’d be like saying all novel internet products have to go through Google, Facebook, and Amazon. And it’s just crazy if you say it that way.
And so how does a biotech bring their innovation all the way to patients? And that means we have to be able to take a drug from conceptualization through commercialization in under 10 years. And specifically 10 years because that’s a VC fund time horizon. Ideally, it’s something closer to seven or eight so that they can actually generate cash flows and we can figure out an exit based on revenue-based fundamentals. And, sorry, if we do that, I think we allow for more founder-led biotechs. We allow for more scientists to go and build these companies. And I think there’s going to be—you know, the cost of capital will necessarily drop if we don’t need to wait 15 years for approval but seven or eight. We bring in more investors who I think sit on the sidelines or are concerned or worried about this because they can think about cash flows at the end of it.
Like, people say tech VCs are scared of biotech, but tech VCs are investing in flying cars and nuclear energy and quantum computers and all this other stuff—they have no problem with it. It’s just like, the problem with biotech is that no biotech CEO thinks about the “P” side of a P&L. And tech VCs, you know, like businesses that make money. That’s how compounders work.
Steve Brotman: It does provide a little bit of a discipline though. It’s hard to hide that you don’t have revenue, whereas in a biotech, you can kind of mush the look. From a non—I’m a non-biotech investor, but I find it super interesting. And as I mentioned earlier, we have a lot of companies that intersect. In fact, my biggest win in my life, in my career—well, early win—was a company called Medidata. And it just shortened—I don’t know if you—it’s electronic data capture. It basically removed accounting and paper from the system. So we saved a year out of the life. So, I can deal with that, but, you know, doing what you do is pretty remarkable. Internally, like, I can’t imagine you guys all are bioscience focused. Is it pretty much you and Joe make a call? Or is it each partner makes their own call in terms of deployment?
Francisco Gimenez: No, we have sort of IC voting structure—investment committee voting structure. So, I think we are very much a team-focused at—you know, there’s a partner that’s going to lead any given decision and investment decision. We have a team that’s meant to make that partner’s decision better. And sometimes they vote against the deal and it doesn’t pass. That’s, you know, tough. But I think the point is not to find ways to poke holes in somebody’s investment but to make sure that we’re finding a way to get to “yes” that is well-articulated. And if there’s disagreement, at least we all agree on the points of disagreement. And, you know, I think some of our best investments have been the most split votes. And I think when I talk to other VCs at different firms, that is—
Steve Brotman: Even if there’s a split vote, you’re comfortable? It doesn’t have to be unanimous. It can be kind of a split vote?
Francisco Gimenez: Yeah, honestly, our best investments have been split votes.
Steve Brotman: You know, a lot—my sister’s, and sorry to call her out—but she’s got her PhD at Johns Hopkins, and she’s impacted by some of the cutbacks, or was worried about some of the cutbacks happening there. You know, do you have a hot take on—is that going to be, you know, Armageddon for the US sciences? It can’t necessarily be great. Is there a silver lining? Is this going to encourage people to get more—is it going to drive more? You know, I was just curious, like, what—from your perspective, from the venture side—how is this impacting your universe?
Francisco Gimenez: So, we’ve written a piece about kind of our recommendations to the NIH. So, I’m just kind of speaking to those points. But I think that there’s clearly a difference between what is stated and sort of the initial position and then what actually manifests. I think that reduction of indirect—and so indirect financing, of course, means that for every dollar that you get as a grant, some percent above that amount covers kind of, you know, basic rent and stuff. And so if I get a dollar of a grant, you know, maybe if there’s like a 50% overhead, then my actual grant is for a dollar fifty. It’s important to think about that because it’s added on top. It’s not 50% of my grant that is the indirect. But it’s—and you have to go into why. Like, the ones with massive indirects, like 87% or 90%, those tend to be academic research centers with hospitals, right? And so those hospitals are paid for via those indirects. It’s not just administrative bloat. There’s a lot of administrative bloat, don’t get me wrong. But when we see these cartoonishly large numbers, we realize there’s a hospital attached to that entity that’s being paid for and paying for a lot of healthcare. It’s like a—it’s an indirect socialized healthcare, if you will.
And so, I think that indirects could be done better. There’s a lot swept under the rug for the sake of just saying it’s easier this way—like a lab never has to think about paying rent because that is swept up in the indirect—or paying for power or whatnot. Whereas if you’re a biotech, you know, you have to pay all of these things explicitly. But the problem is when the indirect is a straight across-the-line percentage, the lab doesn’t get a chance to negotiate that or think about finding cost-saving measures. And maybe, like, of those grant dollars I bring in, how much is coming back to me. It’s almost like federal taxes that go back to states, right? Like, are you a net tax producer or a net tax receiver from the federal government? I think of it that way indirectly.
And so, in general, I kind of view it as more decentralized is better. Reducing indirect costs and making the costs that go into direct and indirect costs explicit indirect costs is probably the right way to go, modular up to some sort of limits here. The idea of cutting the NIH budget by 40% would catastrophically destroy science in America, and it’s insane. But I also think that that was a headline number, and I think in the manifestations of what we’re seeing in the budget, that’s not necessarily happening.
But, you know, the NIH—they have a large budget, but they only have about 10% wiggle room of that budget in any given year because they give out grants on multi-year time horizons. And so if you were to cut it by 10%, that basically has eliminated all your free spending and would just cover the sort of pre-specified costs. But if we go down there, we’re actually cutting into people’s existing grants, which is obviously what’s happening.
So, I think that American innovation is possibly one of the most impressive crown jewels that we have. And so cutting it is crazy. It’d be like saying, like, in the Enlightenment if we decided that we wanted to cut all science spending for one of the most important periods of innovation in human history. And I think the one thing that America has consistently done well is no matter how many mistakes we make on many policy or whatever issues, we always innovate our way out of a paper bag. And so, gutting innovation—and that’s not just NIH, right? Like, this is a message to all other sorts of innovative government sectors—but, and I don’t think our government actually wants that. I think, you know, there’s a bone to pick, and there were mistakes made, certainly. But I think this is a very important sector of our economy, and honestly, just to be a human, right? Like, it’s inspiring to say we live in a country that has the best science in the world.
Steve Brotman: Right. Yeah. So, like you mentioned, this is sort of backdoor socialized medicine now. You know, shouldn’t we just be explicit about that? It’s like, “Hey, here’s a grant. Help poor people,” as opposed to like saying it’s science. Is that like the best way to do it? Is that what you’re saying or—
Francisco Gimenez: Yeah, maybe socialized medicine is too hot of a take, but a hospital, right? Anybody can show up to any hospital and get care. That is, you know, backdoor socialized medicine. That’s not just an academic medical center, right? And they’re ethically not allowed to refuse people because they can’t pay. And I think the added part of academic medical centers is they can do innovative treatments on things that don’t have interventions or in diseases that are considered rare that, you know, there’s not a market demand for. And I think that’s really important to allow for—is that when you go to a CHOP or a Stanford Medical or UCSF, you’re getting the most innovative treatment if you have, you know, a rare disease or if you have a cancer with unmet need that nobody knows how to intervene on. And you can’t just do that anywhere, and there’s no market need for that.
But in curing that, enabling physicians to cure those diseases—or at least try to—we learn about disease biology and fundamental disease biology. Those are the front lines of clinical research, not just biological research. And yeah, like that’s paid for under these indirect costs in NIH. But like, I want that to be paid for. It’s not like we’re giving everybody kind of socialized medicine. It’s just like—
Steve Brotman: Anything you would—you’ve been doing VC for a while now. Anything that you wish you knew when you started out that you know now, that you’d like to share?
Francisco Gimenez: I think that one of the trickiest and most important things to do as a VC is just developing a good eye for great founders—or great eye for great founders. There’s a lot of great scientists, there’s a lot of great programmers, there’s a lot of great—and they start companies and we want them all to do that. But if you’re investing in the early stage, if you just get to choose one skill, it’s identifying great, great, great founders. And that’s just—some people kind of can do it natively. Like, clearly Peter Thiel or Paul Graham, for that matter, just clearly had that eye.
I think I saw a great interview with Sam Altman where he said at YC, they train that in people, and it takes a couple years, but they can—they feel like it’s a trainable skill. And I think that it’s hard to say you could speed-run that, but that is, I think, probably one of the most important reps to the point where it’s like new associates should be spending as much of their time with the best founders in a portfolio, in a fund, and just meeting them and just developing a gestalt for what makes them great and meeting great founders. I think that’ll go probably the longest way.
But also, you know, I spent my whole career up until that point studying AI and technology. So maybe like, if I had a great eye for founders and no idea how to understand technology, I would’ve been saying the opposite.
Steve Brotman: My son’s actually taking his first steps, you know, without his dad’s help in venture, and I’ve been telling him that this is a key skill. And he’s talking to a couple folks. But it’s hard. It’s actually kind of hard. It’s like the hardest part of our job is sort of weighing founders in a way that—you find that edge. And the only way to do it is reps, right? It’s like seeing what works, seeing how they work. And every founder’s different, and they have different edges. Any advice you’d give to a founder?
Francisco Gimenez: Ooh, man. Again, I feel less and less inclined or able to give advice to founders, but—
Steve Brotman: How about founders who want to get in touch with you? The best way to approach you?
Francisco Gimenez: Yeah, I think the oft-quoted—and I still think—best way is warm intro from a person that we strongly respect. There’s a certain category of people who intro us to a founder, and we’re immediately looking for ways to say yes. And I take cold intros and cold pitches. You know, sometimes it’s just—if I have to just sift through the massive prioritization queue, the more you are well-referenced and recommended by people we care about, I guess the more we are trying to find a way to invest in you.
And I think Joe said something really interesting once, which is like, it’s almost like manufacture two orthogonal introductions from people we respect. I think that is maybe too much to myself, but I think it’s just advice we give to founders pitching elsewhere. If two totally different people say, “This is the smartest person I’ve ever met,” you’re probably going to get invested in—at least at the very early stage.
Steve Brotman: Excellent. Thank you so much. Francisco Gimenez, partner at 8VC. Thank you so much for your time today, and I look forward to continuing our chat soon.
Francisco Gimenez: Thanks, Steve.
Steve Brotman: Thanks a lot.
