Chemistry cofounder: “more talented founders than ever” are moving into health care | Term Sheet

15:52Allie:Well, I was going to say, I actually covered your first investment in chemistry, Assort Health. At the time, you said something to me that I've thought about a lot since, that you said you felt like you were being really picky because those first couple of deals were going to be a signal. Assort has since actually ripped, right? There's a long way to go, but they've raised another round. And it seems by all accounts, they're doing great. Do you still feel like you're being picky? What does that look like for you?

16:19Kristina Shen:Oh, you're so good at picking my words from earlier in the year. This is the danger of talking to me over and over, Christina. Good memory. Well, first of all, all the credit to Assort Health Success all goes to John and Jeff. I mean, I think they are phenomenal founders. And I think from day one, what excited me about them was that they always have a very high bar. And they were this rare combination of people who are technical, iterate really quickly, and they really cared. And I think that has translated to them building a phenomenal company in the healthcare voice space. I think to your question on me being picky, I mean, you know what?

16:56Kristina Shen:I think I deserve to be picky. That's always been my opinion. I think I continue to be very picky because I think that while we're in a really interesting time in AI where there's going to be lots of value creation, I think that I've always wanted to work with the best. And I think just like when founders hire the best talent to maintain a high bar so that other employees look around and say, oh, I want to work with these people. I think the same is true for building chemistry's brand, which is we only want to invest in the best. I think it compounds our brand. And hopefully that means it compounds our returns as well.

17:34Allie:Well, one of the things I found fascinating, too, is I've been doing in part a sort came to me because they emailed me a bunch of times and you guys emailed me a bunch of times. Persistent matters. Persistence absolutely does matter. And I was glad I said yes. But it was also the sort of thing I became very interested in this sort of intersection of AI and health care because there's so much investment going into the space right now. Are you looking at other AI and healthcare deals? How are you thinking about longevity there, especially because it's such a complex industry?

18:02Kristina Shen:We are actually spending a lot of time in healthcare. And quite frankly, I think something like a quarter of the investments that we've made to date have touched healthcare in some fashion. Oh, so you're in it. I think there's a bunch of reasons why that's happened. So one, I think we follow where founder talent is going. And I think smarter and more talented founders than ever are starting to move into healthcare for a bunch of reasons. It's a massive market. The second being I think AI in particular translates really well in healthcare. Why? Historically, healthcare has been very labor intensive, back office heavy, lots of repetitive tasks.

18:41Kristina Shen:Those are all components that really cater well for AI and LLMs. You know, in particular, I think software historically has been harder in healthcare. and I spent a lot of time in software. I've been investing in software for over 15 years because software, you have to sell productivity to a human. Well, now with AI, you're finally able to displace certain humans in the back office. It's why we've seen AI be so successful displacing BPO's. So for those at home, wait a second. What is BPO? Sorry, business process automation. So services industries where you outsource some capability to an outsource party to handle.

19:13Kristina Shen:A lot of times it's been call centers, a lot of times it's been billing, like in healthcare and healthcare codes. And in a Swarth Health's case, it's been outsourced call centers. Right. So what a Swarth Health does is they use voice agents to enable patients to schedule appointments with their doctors at any time.

19:30Allie:Like if I called them right now to a doctor in New York, I would be able to schedule an appointment.

19:35Kristina Shen:You could call at 3.32 a.m. and get hold of any doctor's agent that you need to in order to make a really valuable appointment. Which actually, like, we say it a little bit off the cuff now because we just expect that voice agents can do anything now. But you have to remember how important that is, right? You know, for some of these specialties that a SWORD Health sells into, sometimes people are waiting weeks, if not months, to get really important care for either themselves or their family members. And so now with AI and particularly our SOAR Health and healthcare, you're able to make those appointments and actually not only help the patient, but also help the doctor office in terms of displacing a lot of costs.

20:17Allie:Where wouldn't you invest at the intersection of AI and healthcare right now? Where are you like, absolutely not, I don't think that's going to work or it shouldn't work?

20:25Kristina Shen:Ooh, that's such a good question. Okay, so I'll start with where I would invest. I think there's a lot of back office costs. I think AI can be really productive in displacing costs. I actually am really interested in consumer health care these days. There are many things that can mean. Like, are we talking wearables? Well, OK, yeah, I think it mean a lot of things. I think that the health care system in the U.S. historically has been so hard to navigate that consumers are taking control of their health care. And so what does that mean? There's new reimbursement codes that enable consumers to get certain health care benefits or services without having to pay for them directly because CMS codes allow you to.

21:03I think that consumers are willing to pay out of pocket to get certain advisory help or understand how to help either their parents or themselves.

21:13Kristina Shen:I'm really into a lot of those components, which is I think consumers want control of their health care. And I think they're going to get it out.

21:18Allie:There's so much evidence like the rise of wearables, Oura rings, blood tests, right? Like functional health, right? Exactly. I think there's a very interesting, increasingly personalized future of health care, especially because the system is so inefficient. People want to feel like they have agency. Exactly.

21:34Kristina Shen:And I think, you know, Function Health is a great example of consumers are willing to pay a significant amount of money out of pocket in order to get more data on their health. And I think that data is going to be the basis of a lot of interesting new companies as well, whether pharmaceuticals build drugs around it or et cetera. There's going to be a lot of directions they can go. So then your question is, where would I not invest? You know, I still think that a doctor is important. And so while I think there will be huge transformations in how you can both help doctors make better decisions or help consumers better educated as their options, I believe there is a place for a doctor.

22:11Kristina Shen:And so direct displacement is probably not where I'm focused.

22:13Allie:I'm going to tell you about something I worry about. Oh, yes, please. I think all the time about the wave of startups coming into health care, you know, a system that is already so strained. And I worry about diligence on some of these companies, right, especially the ones that get to the place where they're interfacing with patients. And I worry about the bubble pop, too, and how it might affect some of those companies. How do you think about those parts of the equation?

22:38Kristina Shen:I mean, I think the responsibility piece is very real. I would hope that the efficiency of capital means that we're trying to back founders that want to make a change in the world. But I think it's going to be, it's hard to control. And I think that's going to be one thing that we have to be cautious as both a tech and venture ecosystem to be careful with. The bubble burst, I think, is a real good question. I think there's no denying we're in a very frothy part of the market. I think valuations are high. We have the comparison to 2021 all the time. However, I very much believe the next generation of Googles, Facebooks are going to be created from this platform shift.

23:19Kristina Shen:And so while we are going to see some bubble bursts at some point, I believe we're going to see tremendous value creation as well.

23:27Allie:What parts of the market do you think are bubbly and which ones do you think have staying power?

23:32Kristina Shen:I think there's a lot of talk around quality of revenue these days. And why are we still using SaaS era ARR? Somebody explain it to me. You know, and it's even more extreme than that, which is, you know, there's AR, there's car or contracted AR, then there's car from two, three years out. I think it's important to understand what metrics are relevant to the specific business. But I think it's OK that some companies have really high top of funnel from great consumers trying their products. And they might churn out and then there's some stable base. I care a lot about what is revenue in six months, right?

24:13Kristina Shen:Not what is revenue at the point in time today. A lot of people are mixing, you know, month one times 12, which I think is complicated. I see what you're doing. And 12 months is a long time. 12 months is a long time for some of these companies, given that many of them have only been around for less than 12 months. Look, I have candy bars older than some of these companies, probably. For sure. I think that somewhere down the line in the next three, six, 12 months, I think there is going to be a little bit of reckoning of what real revenue is and what is fantastic top of funnel interest, experimentation, consumer revenue.

24:50Kristina Shen:But that doesn't mean it's a bad thing. Like, I think there's also a lot written about low quality revenue. And I don't think, well, I think not all revenue is created equal. I think having a phenomenal top of funnel is hard in this industry. You want distribution access. You want customers to want to come to you. And then it's up to founders and companies to build enduring products that are sustainable.

25:12Allie:I was going to say, can low quality revenue become high quality revenue? Is that possible? Hmm, that's a great question.

25:20Kristina Shen:Okay, I think we're asking the wrong question. I think the right question is, to your buyer, are you the most important software or most important AI product to them? And if you're not number one or number two on the list, you probably fall below the line. And I think that is the test that we as investors and founders should be asking themselves. It's fantastic if you're growing from zero to 10 million in year one. But if to your customer, you are the 10th product, I think there's going to be some sustainability questions. And that's what we ask. So we do a tremendous number of customer calls.

25:56Kristina Shen:And all we do is talk to end customers. And that's what we ask them. Is this the number one product for you? And if it is, that's meaningful. If you're number five, that's hard.

26:04Allie:Now tell me a little bit about asking for a friend. How will we, you, me, all of us know when the reckoning has started? Or will we know? Well, if you hear the reckoning first, I hope you cue me in.

26:18Kristina Shen:I'll call you, you call me. I think there is very little value at the early stage to trying to decide when the reckoning is. I mean, think about the last bull market. People called it in 2015, 2017. Finally, we had a little bounce in 2020. And then it obviously, actually investing 2020 was actually a phenomenal year to invest. So I don't think anyone knows when the reckoning is going to hit. I think it will start in the public markets. And everything takes a quarter to roll its way down to the early stage. And so it's like the public markets that translates to market value, then to late stage private, then to early stage.

26:56Kristina Shen:I think there's a lot of money in the venture capital ecosystem. And so I think it's more about making sure you're working with the longstanding companies that are going to be generational companies as opposed to trying to decide when the reckoning hits.

27:10Allie:I was going to say, as we sort of wind down here, one of the things that gets talked about a lot are the incumbents, whether it's the metas and the Googles, the open AIs, the anthropics, which I think we can call them incumbents now. I don't I don't even know. Where do you think the balance of power lies right now? Who's at the top of their game?

27:30Kristina Shen:First of all, I think the league tables are constantly changing, even in the current AI lab world. Gemini just released a phenomenal model that has done very well against the benchmarks. And OpenAI is reportedly in a code red. And OpenAI is in a code red. And by the way, actually, to me, for what it's worth, I think that signals that everyone knows that they have to be on their toes. I actually don't view that as a signal weakness. I view that as OpenAI understands that we are still so early in this AI platform race that they know they have to prioritize. So I think the lead tables are going to change multiple times over the next couple of years.

28:09Kristina Shen:I think Google has put forth a fantastic showing. I think Anthropic is rumored to go public and have done a really good job, particularly with Cloud Code and entering into the enterprise. And what I really think is that each lab is going to become very well known for something that they're really good at. And it's very possible that they bifurcate the market into whatever it's going to be, code, consumer. You look at Google and Microsoft, they have tremendous distribution benefits into the enterprise. I think we're still in such early stages that they're going to switch multiple times.

28:43Allie:The thing I love about the league tables analogy is because it gives me the image of European soccer where companies or see I'm even saying companies teams are relegated like they can fall down a league. And it sounds like what you're saying is that actually can and does happen here.

28:58Kristina Shen:I think it very much will happen. And we might be pleasantly surprised and we haven't even talked about the Chinese labs and how they're going to fall into league tables and how interoperable dynamic.

29:06Allie:That's so underreported on in the U.S., by the way. I mean, someone actually said to me today, they were like, you know what the real code red is? China. China's AI in China.

29:15Kristina Shen:I mean, if you look at a lot of what's going on in the image and video generation world, I think the Chinese models have had tremendous distribution. One of our portfolio companies, Comfy UI, they basically build AI workflows for images and video. A big percentage of the models that are being leveraged by consumers are Chinese models. And so, you know, I think we're really early in this competition.

29:39Allie:To wrap this up here, Christina, give me one prediction for 2026 so we can look back on it in a year and I can tell you what you said. One prediction.

29:49Kristina Shen:All right. So my prediction for 2026, I think it's going to be the year that computer use has tremendous applications at the application layer. And I'll give you a little context on this. So I always look at the AI labs as early signal as what the infrastructure breakthroughs are going to be and see what happens at that layer. So I think 2024 was all about cogeneration, produce cognition and cursor. I would say 2025 has been a year of cogeneration. Yes, exactly. I would say this past year has been all about speech-to-speech models and voice. And we've seen companies like Assort Health, but also companies like Sierra and Decagon emerge in customer support.

30:26Kristina Shen:And I think computer use is going to be tremendous in terms of agents being able to control how work is done, particularly in legacy industries. and in industries where there's not a lot of APIs and connected software and integrations to get work done. I think computer use is going to enable legacy industries like manufacturing, logistics, healthcare, stodgy historical industries get a lot of back office work done

30:51Allie:and eat a lot of share from what we've historically called the RPA industry. So for some reason, when you say computer use, the first thing I think of is sitting in a basement with a giant dinosaur of a computer. I don't know why. I think it's because I say laptops, But it sounds like that is actually in part kind of what you're talking about, some of those tasks that people normally would sit down and do on a computer.

31:09Kristina Shen:Exactly. I mean, these very repetitive tasks that get outsourced again to BPO, business process automation players, where you've got huge amounts of labor manually inputting data into ERPs, completing a form, answering emails. I think computer use agents, which has been the next breakthrough from the AI labs, are going to transform all of those industries and enable the automation of a lot of repetitive tasks that we see across services industries today. Okay.

31:41Allie:Chemistry is a really interesting word in the context of startups. What fells a lot of early stage businesses, what fells a lot of venture firms is a failure of compatibility, a failure of chemistry. In your experience, what gives a team chemistry?

31:55Kristina Shen:Such a good question, because obviously we care a lot about chemistry. We named ourself chemistry because we care about the interpersonal dynamics between the three partners, as well as our chemistry with founders. And so what does chemistry mean to me? Honestly, I think it means trust, right? Because when you have great chemistry, and we see this between the three of us, as well as chemistry between co-founders, for example, when we're investing in a team, it's about do we believe that they can run quickly and iterate quickly because they trust each other? That's what we're really looking for.

32:31Kristina Shen:And when you have chemistry or trust, you can really achieve anything. But if your team doesn't have chemistry, you could imagine co-founders bicker over small things. They are misaligned on what their end goals are. But if you have that chemistry or trust, I think you can really do anything. How can you tell if trust has been built? I actually think it's in all the small things. So one thing that I really love doing before we invest in any team, we love spending in-person time with them. We really love grabbing a dinner or a meal with the co-founders. And it's because you can see them finishing each other's sentences.

33:06Kristina Shen:You can see how they allow each other to answer one question versus the other and not talk over each other. All those little signals are the little chemistry pieces that you pick up that are the makings of trust. And we're at the early stages. I mean, who knows if our theses are right or wrong. Or the company's a completely different company in five years versus months. Things of all, companies pivot. it, we're backing people. And we're backing people who have a sparkle of an idea, great chemistry between them, and a passion for what they're doing. And I think if we can get all those right in our early founders, I think we'll have a phenomenal group.

33:46Allie:One of the greatest risks at any of these early stage companies, right, is a co-founder breakup. How can you tell if co-founders are going to stay together? What are the red flags you look for in those relationships?

33:57Kristina Shen:The honest truth is we don't really know, but there's little signals that we really cue in on. So I really look at the division of labor. So I think great co-founders understand I'm good at X, you're good at Y, and we let each other run wild at what we're good at, but still come back to North Star. That's the second thing I look at, which is many times I meet co-founders one-on-one. It's not always a common thing. I meet them one-on-one frequently, And it's because I want to hear that their North Star is the same North Star. What do they want this company to become? You want to hear they're speaking the same language.

34:32Kristina Shen:Exactly. Because quite frankly, like, and I learned a lot about co-founders from even Mark, Ethan, and myself coming together. We're co-founders. We had to get aligned. We had to know what we wanted chemistry to be today. And we wanted chemistry to be 5, 10, 15, 20 years down the line. Hopefully it outlives all of us. And I think it's probably one of the most underrated things that people forget about is that you have to be aligned as to what you want the future to be. And it's the number one test with co-founders as well for founding a company is do they want the same thing?

35:04Allie:Mark and Ethan are your first co-founders. Yes. What did coming together with Mark and Ethan teach you about co-founder relationships and about chemistry? Well, I mean, gosh, I've learned so much about myself personally as well as each other.

35:17Kristina Shen:You know, we did everything, which is I think we did the right thing, which we focused on. What do we care about in the future and we want chemistry to become before we figured out all the little strategic details of even what chemistry would be called or what the strategy was? I remember so distinctly that when I first left Andreessen, the first person I told was like, OK, I'm going to move on from Andreessen. I'm going to start a fund. And they go, oh, so who's your LP? And I was like, oh, I don't have one yet. It's like, ooh, are you sure that was a good decision?

35:49Allie:And I first freaked out. Whoever that is, look at you now.

35:53Kristina Shen:Well, I don't know about that, but I first freaked out. And then I really like, no, you're wrong. Because the tactics and the operational components, we can figure out. But if we can't figure out if the three of us should come together and we want the same thing for chemistry, then that is the most important thing ever. And I think I advise so many people who are starting funds. and it's the part that I think people kind of jump over because they get so excited about the tactics and the strategy.

36:19Allie:Well, in the end, I think you have to be aligned together because it's a very long journey together. It's funny. I'm remembering something right now. Back when I met Jeff and John at a sort, I asked how they came together and they did a very extensive co-founder questionnaire. I believe it is originally from first round. And it's the sort of thing where they basically sat around and we're like, here are all of the things about me. And I think there's really a lot of value in that because I mean, it's like a marriage. It is like a marriage. And like, it's actually

36:48Kristina Shen:probably harder to get divorced from your co founder. It is actually logistically more difficult. It's very difficult. I say the same thing with founders, too. It's actually harder to get rid of your board member for someone that you work with than to actually get divorced. I think it's important. You know, we did everything from we shared each other's performance reviews, which had all the good and all the bad in there. We kind of did 360s on each other. We spent a lot of soak time, all three of us together, two people like one on one in each combination. We had a triple date, which was very important with all the plus ones, because you do know the spouses do matter.

37:27Kristina Shen:But I think the same is true for co-founders starting a company, right? Like, you got to be aligned from every component or respect every component of each other's lives, even if you might not be identically the same. And I think that is all tied to having the same

37:44Allie:North Star. We are at time, but I do want to ask you one more question, if that's okay. For the founders listening to this, what is one piece of advice you think every founder needs to hear? And what would you say if you weren't afraid of hurting anyone's feelings? Thanks.

37:58Kristina Shen:Ew. Okay. Let's see. Of course, you always get the zinger right in the end. Look, my biggest biggest advice, honestly, is trust your gut. I mean, I think everyone's intuition is generally right. Trust your gut around, do listen to the advice that you're getting. Trust your gut around the people that you really want to work with. Trust your gut around, does this feel like a sustainable trend or not a sustainable trend? And I think the best founders ignore everybody and trust what they believe. And that's how I operate in my life. You know, I've gotten a tremendous amount of advice. Some of it was great.

38:35Kristina Shen:Some of it was not so great. But in the end, you know, I think I've been right. And so I, maybe that's not how I want to end this conversation.

38:42Allie:But no, I actually, I love it. How often do we have a woman just sit there and say unequivocally, I think I've been right. I need some of that energy. I'm tired of people telling me that, you know, I wasn't right.

38:51Kristina Shen:And so, like, I'm not always right, but I think I will never regret trusting and doing what I believe. And so that's maybe the takeaway I'd say is, like, trust your gut and no one's going to know what's right for you better than yourself.

39:04Allie:Christina, thank you so much. Thank you so much for having me. This is great. And that was Christina. I keep thinking about what Christina said about the league tables needing to change in venture and how AI will ultimately change who can win in venture. We'll have to see if that's ultimately true on a firm level. But I do think there is something that she's saying that is fundamentally the case, which is that this next era is absolutely going to be about speed in one way or another. It is going to be about alignment and it is going to be about believing before it's obvious. Now, you can say that's always true of venture, but my joke about venture capitalists is that there are a world of contrarians who are remarkably good at following one another.

39:41Allie:And it's important to remember that maybe for this next era, that won't necessarily be the case. That's it for Term Sheet. See you next week.

39:52Allie:Termsheet is a Fortune magazine podcast. Our producers are Joyce Coe and Chase McCleary. Our production manager is Sam Freund. Our supervising producer is Shako Liu. Our executive producer is Lydia Randall. Fortune's head of video is Adam Banneke.

40:08Allie:30 years ago, Blinds.com broke the mold and made custom window treatments easy for everyone. Over 25 million windows later, we're celebrating by giving our customers up to 50 % off site-wide during our anniversary sale. Whether you DIY it or want a pro to handle everything from measure to install, Blinds.com has you covered. Shop online, access real design professionals, and get free samples. Thank you for 30 amazing years. Shop the anniversary sale now through March 11th and get up to 50 % off site-wide at Blinds.com.