Vinay's Journey: Mastering AI & VC from Harvard to Bessemer, Two Sigma, & Foundation Capital | EP 22
Description
In this episode of UNinvested, Sahil interviews Vinay Iyengar, a seasoned venture capitalist with a background from Harvard, Vesemir Venture Partners, Two Sigma, and Foundation Capital. Vinay shares insights on the venture capital landscape, the importance of working in smaller organizations, and the evolving role of AI in startups. He also discusses his personal investment strategies and the balance between passion and practicality in entrepreneurship.
What we explore:
Vinay's career journey from Harvard to Vesemir Venture Partners, Two Sigma, and Foundation Capital
The impact of working in large vs. small organizations
The role of venture capital in early-stage startups
The democratization of technology and its effects on non-technical founders
Lean startup methodology and the importance of iteration
Balancing passion with practicality in entrepreneurial ventures
The potential risks and benefits of AI in various industries
Vinay's approach to angel investing and supporting friends' ventures
The importance of routines and personal relationships in maintaining motivation
Where to find Uninvested:
Website: https://www.uninvested.org/
In this episode, we cover:
[00:00:00] Introduction and welcome to Vinay Iyengar
[00:00:08] Vinay's background and career trajectory
[00:00:37] Impact of working at larger organizations vs. smaller ones
[00:01:03] Transition from Vesemir to Two Sigma and entrepreneurial motivations
[00:01:46] Accelerating one's career by joining growing organizations
[00:02:00] Foundation Capital's focus on early-stage startups
[00:02:21] Building a thesis around specific spaces and technologies
[00:02:57] Role of generative AI and founder conviction in investments
[00:04:11] The balance between founder-driven and VC-driven theses
[00:04:55] Challenges and opportunities for non-technical founders
[00:06:12] The democratization of creation and the rise of no-code tools
[00:07:05] Importance of founders building products themselves
[00:08:00] Lean startup methodology and the importance of iteration
[00:09:00] Deciding when to move on from a venture
[00:10:02] Passion for the idea vs. passion for building
[00:11:32] Insights from personal investments and supporting friends
[00:13:00] AI's potential risks and societal impacts
[00:17:00] AI in security and the rise of phishing and smishing attacks
[00:21:00] Service as a software and future trends in AI
[00:24:00] High-stakes interactions and the need for human oversight
[00:25:00] Importance of personal routines and relationships
[00:26:06] Closing remarks and thank you to Vinay
[00:00:00] Sahil: Welcome back to UNinvested, where we break down everything entrepreneurship and tech so anyone can understand. I'm Sahil and today we'll be interviewing Vinay Iyengar. Vinay has an impressive background, starting at Harvard before working in arguably the most storied venture capital firm, Bessemer Venture Partners.
Since then, he's gone on to work at Two Sigma, Foundation Capital, where he focuses on very early stage companies. All this, alongside a plethora of angel investors. Welcome Vinay One trend , looking at your background You went to Harvard, to Bessemer, to Two Sigma Capital, to Foundation Capital.
It seems like you kind of go from bigger organizations to smaller ones.
[00:00:37] Vinay: I think you can have more impact at smaller organizations. And so I think early in one's career, it makes a lot of sense to work at places that are large and established, And that, you know, have a track record of training people and teaching people.
But once you feel like you've learned all that you can learn I'm the type of person at least who, who has always enjoyed trying to take a bet on myself whenever possible. And so I think that was sort of the reason, especially when I think about the transition from Bessemer to DoSigma, that was very much coming from my own sort of entrepreneurial itches and my desire to go try and build something new from the ground up.
[00:01:08] Sahil: I want to push back on that. Like when you're at a big organization, You might have more reach, more resources to actually make a bigger impact.
[00:01:15] Vinay: Yeah.
I mean, I think that's fair to an extent. I think that's probably specifically true if you're in a more senior role or you're a decision maker at a very large organization, but as someone early in their career, it's, it's, you know, you can't just get a job running a large organization or business, whether it's a large venture fund or a large company or startup.
And so I've always thought that You know, the quickest way to accelerate one's career and get to one of those roles where you're actually be a decision maker is to join an organization that's small and growing and grow with them and help turn it into something big. And by the time it's big, hopefully you're in a role where you can actually have some surreal impact
[00:01:49] Sahil: right now at foundation capital, you're making the most impact you've had in your career.
[00:01:54] Vinay: Definitely don't know that. I can't say that for sure. I think. Every day I do get to work with founders really closely. And you know, in my role as a board member and advisor to companies, I definitely feel like I, I can have some impact, but you know, probably again, more than I had five years ago, but I still have a long, long way to go and a lot more impact that I want to have in the world.
So you know, I, I'm very much still just getting started.
[00:02:18] Sahil: You guys really focus on early, early stage startups before That quote of, you know, the death valley of startups where you don't have any customers. You're really just an idea of you invest based off founder conviction.
[00:02:29] Vinay: It's, it's a few things, right?
One, I think we believe very strongly in focus and going very, very deep, building a thesis around a space. As an example, now I'm spending a ton of time in and around all things generative AI. And I think part of the, part of the idea is like, if you meet hundreds and hundreds of companies that are focused on a very specific problem area or a very specific type of technology, then the next founder you meet, even if there isn't sort of clear traction by way of metrics or Your product, you sort of have a better intuition of what's a, what's a good idea or what isn't.
So I think that's, that's one thing is like when you, when you go really early and you're very focused, that helps you drive better decision making. But then to your point, and the second thing is really making bets on teams and trying to find teams that are truly exceptional and extraordinary in some way.
[00:03:19] Sahil: So you kind of have a thesis and then go find the founders, like within your thesis area. Has it ever been on a scenario where the other way around has happened? And you've found it. Founder that was outside of your thesis.
[00:03:28] Vinay: Yeah, it's, it's definitely happened many times. I would say you know, part of the advantage also of, of going very early is you could actually shape that founder's thesis building as well.
And your pattern recognition and all the previous conversations you have with founders actually help you sort of almost co create with that founder. So at the stage where invested, we almost like to think of ourselves as like a co founder. And certainly we wouldn't, we, we are not taking that level of credit and it's really founders that drives great outcomes.
But we like to think of it as like a melding of the PCs that ultimately helps us really build big and exciting companies. Yeah, I, I mean, I would say to your direct question, like who's really driving the pieces, the founder of the VC. I would say it's almost always the founder. Actually, we can come up with a high level perspective of like, Hey, I think this market is very interesting or this technology is interesting here are the ways that we, we believe you can commercialize it here are the go to market strategies that are effective, but the actual like core insight around a startup, I believe always has to come from a founder and from their unique experience, it's, it's very hard for us to come up with that and then go find the founder.
So, so when I say we're thesis driven, what I really mean by that is we, we follow high level themes and trends and patterns. But the actual idea behind like great startups, I think always comes from like brilliant entrepreneurs.
[00:04:52] Sahil: You mentioned people building startups around their insights, probably typically like an experience they had in the industry, but what do you do?
Like you come across a founder, you love his idea, but he necessarily doesn't have the skills or. The internal team himself to put it to fruition.
[00:05:08] Vinay: Entering a world where the ability to build technology is slowly being democratized, right? You have a bunch of these. Low code, no code tools that exist whether things like retool or whatever that help folks who are not builders themselves build and iterate with an idea, I think also, you know, it's increasingly easy to sell a customer on like a prototype or a demo or an MVP that isn't actually the product, but sort of like a, almost just the front end of what a product looks like.
And people can use tools like Figma or Canva to design that create an experience that feels like you're using the product. So I think. You know, maybe 10 or 20 years ago, I think it was very hard for a founder to be non technical and for them to go out and, and sort of really build a very, very large technology company.
But I think over time we've seen the playing field level. And I like to call this trend, like the democratization of creation, like this idea that, you know, it's never been easier to build And to build products that are really like robust and feature rich, right? Because you also have this whole ecosystem of like third party APIs and libraries, things like Stripe Twilio, you know, Shopify, if you're trying to build it, some sort of e commerce business, right?
So you have a bunch of these things that now just make it a lot easier to build. I do tend to believe that technical founders probably outperformed just because their product velocity can be a lot faster, right? So they can. Experiment and iterate probably more quickly, but that certainly doesn't mean it's impossible for non technical founder in today's world to, to build a really big and exciting business.
[00:06:42] Sahil: Going off that, would you want the founder to be using the no code, no code tools themselves? You know, nowadays you can do a plethora of outsourcing, you know, platforms. They had someone cook you up a. MVP maybe just to show up with the concept
[00:06:54] Vinay: again, my bias would be towards people who actually build it themselves because I think just the act of like building and designing a product is just so important for an entrepreneur to do because like it, it's almost like so much gets lost in translation if you muck something up and then give it to a third party agency to try and build like.
You're not really building what the user really wants, and you're not, you're not able to iterate and experiment as fast. And again, I think the most important thing in the early days of companies is like, your pace of experimentation needs to be very, very high. And so the only way to do that is for you, the founder or CEO, to actually be in the loop.
During that actual product creation process.
[00:07:37] Sahil: We're talking about like the lean startup model of what YC preaches, iterate fast, ship you a version of all the time, kind of balance that with doing heavy research, you know, design research before you even create a product.
[00:07:48] Vinay: I am a big believer in what people call the lean startup methodology, which is like the idea that you should build as little as possible and talk to users as much as possible in the early days of the startup.
And so. What that means is, I think it's a little bit of both of what you were saying, right? It involves Not really like writing a single line of code, but, but trying to build an experiment and create things and put them in front of users that are like truly the minimum viable product and try and get their feedback and try and understand what they use this, what they pay for this, is this solving their pain point and then going back to the drawing board, iterating, talking to more users.
And so I think this continuous loop of like building the minimal viable product and talking to as many users as possible. And, and. You know, doing that virtuous cycle as much as you can. I think that is the playbook for building a great company. And again, I think there are, for every rule, there are exceptions.
There are, there are ways to do this that are, are totally different, but the, the, the method I've seen work the best is, is sort of what I just described.
[00:08:55] Sahil: You yourself, it's on your LinkedIn. You said you had a failed venture. When is it time for a founder to give up or maybe move on to the next idea?
[00:09:03] Vinay: My answer to this is pretty simple.
When the founders lose passion and hunger and, and just lose interest, like that's, that's a good time to stop. But as long as someone is still like really passionate and interested and excited, like, you know, sometimes these things take years and years or a lot of persistence, but it's the true believers and the true optimists who ultimately persevere and build things that really matter.
In my case, like I, that just wasn't me. Like at the time I was, I wasn't even full time on it. I was still a student. You know, realize pretty quickly, like I'm interested in a whole bunch of other things. And, and while this is exciting, I was getting distracted and it, it just, it wasn't something I'd loved at the time.
And so I realized, Hey, I got to move on. And so I think when people are motivated by this fundamental love, this fundamental desire to solve a problem, like again, that's when they'll be most persistent and when they'll really go on to build something that's really interesting. And in my case, I think it was clear that I.
at the time probably wasn't doing it for the right reasons and didn't have that, that love and passion. And so I'm glad I, I sort of cut it off when I did and, and moved on to other things.
[00:10:14] Sahil: So is the love and passion, would you say, have to be for the idea of what you're trying to solve, or is it for building, for trying to have, like, the company look?
[00:10:22] Vinay: Yeah, that's an awesome question. And it's something I think about a lot. And I think my, my opinion on that has changed over time. I think I historically used to think, like, the best companies, the founders need to have, like, this deep love of the domain. Deep with the problem that they're solving. Like they have to just like be obsessed and that's the only way to build a great company.
But then you look at examples of, of great companies that have been built by really passionate young people who aren't necessarily excited about a specific domain, but are just excited about the idea of building something that matters. I think Brex might be like the best and most famous example of this.
You know, a couple of folks who in YC who were just like, Pivoting around, trying a bunch of different things, found this idea that really worked and went for it. And you know, I've heard this now many times of like successful companies that start really because the founders just want to be entrepreneurs and no other reason other than like passion to build and be entrepreneurs.
And they pivot a bunch and experiment with a bunch of different ideas. And finally something sticks and they go on and build something that's like very big and exciting. And so. I think my, my opinion on this is sort of evolved over time.
[00:11:36] Sahil: This lean startup model to constantly tell people to iterate, iterate leads them away from their original idea.
I always think if you're stuck to one domain, you're solid, but that if I want issue, you might miss a bigger opportunity.
[00:11:48] Vinay: Absolutely. Absolutely.
And I think through this process of even talking to users, you discover new pain points, new problems, new markets. And so, and I think. You know, in many ways you actually have an advantage being an outsider and approaching a market or a problem without being a domain expert or someone who comes from that industry, you're an outsider and you have new ideas and new ways of innovating and you aren't like bogged down or stuck by like traditions or Oh, we should do it this way because it's always been done this way.
You've been sort of approach things with a fresh, fresh mind. And I think that's, that's valuable.
[00:12:27] Sahil: I saw you have a lot of personal investments as well. So do you treat those any differently than you do when you're investing within a firm? Or is it kind of, you carry the same mindset, same thesis. You look out for the same things.
[00:12:39] Vinay: I think my philosophy with the angel stuff has been. A few things. One, it's, you know, my, my focus as a VC is fairly narrow. I, I spend most of my time in B2B software. You know, spending a lot of time these days, again, in all things AI. But there are a bunch of other things that I'm like, super interested in and excited about.
And so it's a chance for me to one, like, just play around with these new markets and new ideas that I think are really interesting. An example of that is I, I. Was a very early check in a company called Backbone, which makes a controller that snaps onto your iPhone and basically turns it into a gaming device.
I'm not like a big gamer, but I think gaming is like a fascinating market and a massive one. And so it was, you know, lucky to put an early check into that company. They're doing really well. And so, so one is chance to learn about new markets and new opportunities. And then secondly, it's a chance to support friends who are starting companies who I believe in and I'm going to help them anyways, because they're my friends and I care about them.
And so I'm like, if you're going to, if you're going to be successful and I'm going to help you, I might as well, like benefit if, if, if things go well, and so that's sort of the second, second bucket for me is just a chance to support friends and, and helping them achieve their dreams and, and, And so for the most part, a lot of my angel investments are sort of small checks early on into a company.
I'm probably not as hands on with them as I am with the actual investments that I make from the fund. Because again, with, with a fund, you're, we're investing millions of dollars, we're owning hopefully at least double digits percent, you know, of, of these companies. Often we'll have a board seat, we have an obligation to be really involved.
With the angel checks, it's like small, small amounts into, into early rounds. And and so it's, it's a different sort of risk profile and different involvement, but you know, try to be helpful wherever I can't be.
[00:14:24] Sahil: But so is there any one where you feel, you know, very tied to the mission, you know, not just because you think it's cool, but maybe the product wasn't even there yet, you might've invested way too early, but you were like, I really believe in.
What they're trying to solve.
[00:14:37] Vinay: I'm also reminded of this because I just spoke with a founder yesterday who I put an angel check into who I'm a huge fan of, and it's a company called reflex AI. And basically what they're doing is they're using a lot of the latest advancements in generative AI. to actually train and do quality assurance for crisis counselors and crisis response people.
So an example being like folks who are operating 911 lines or, or crisis lines for the VA or any of these nonprofits these people are dealing with really like intense life and death situations and often don't have good ways to train those operators. And so now we can, we have these gen AI models that can actually.
Sort of help train people more effectively, but also like QA, those conversations in real time, because we now have sort of AI that has a lot more intelligence around conversations. So that's going to be, I'm like super excited about because, you know, it's just it's a, it's a, One of those interesting areas where some of the newest advances in AI is having like a clear social impact on the world and clearly making people better and hopefully actually saving lives in the process.
[00:15:46] Sahil: Do you think there are products that could be hurt with the incorporation of AI? Just, you know, startups just trying to throw an AI aspect in there.
[00:15:53] Vinay: I mean, I think AI generally speaking has a lot of risks and a lot of downsides. I mean, one just, I, I'm, I'm not an economist by training, but like, There's a world in which some of this generative AI stuff really has a meaningful impact on jobs and like automates a lot of jobs away.
And I think that can have probably really negative economic impacts. I think at the same time, like, I'm also very bullish on, on the power of AI and new businesses to create new jobs and new opportunities and ultimately grow the economy. But, but the jury's out on that. You know, secondly, I think AI also has a lot of You know, issues with bias and with you know, there's this very famous example of a chat bot that Microsoft created that ultimately ended up spewing out a bunch of racist, misogynistic, homophobic content because it's trained on the human data and ultimately like devolved into this very scary thing.
And so I think like, you know, there are a lot of risks in AI. Also, these, a lot of these models operate as black boxes today. We don't really know how they work. We don't really know why they're making the decisions that they're making. And so you know, if people start to trust AI too much and don't have a, a good sense of how exactly it's, it's coming up with the answers it's coming up with, like, I think that poses a lot of risk to society.
So there are a huge number of like issues that I think are, are out there with AI. But at the same time, again, I think it promises to be really transformative and, and can, like, push us forward and in, in really meaningful ways. And so it's about this question of, like, how do we develop AI in a way that's responsible and ethical and transparent?
And if we can do that, I think, I think we're, we're in for a very bright future. So what do you think, more so, like, the dangers? I mean, I, I think there are a bunch of, Issues and threats that this new AI stuff poses and to your point about security I think one of the biggest things we're actually seeing data around is the the rise of these phishing and smishing attacks So basically, you know I can now train a model that sounds just like Sahil's voice and I can call up your mom and Make it seem like you got kidnapped and exploit her for money, right?
Like, like the, the threats here are pretty endless. And even, I, I'm sure you've noticed this on your phone. Like you were getting more and more of these like SMS bot hacks that, yeah, that are, you know, basically just phishing schemes that are trying to get you to divulge some sort of private information such that people can hack you.
And so that's just one of the many, many risks that, that you know, This AI stuff introduces, not, not to mention just the fact that now we have a bunch of systems that are just automated chat bots, seem like humans. If those get hacked or exploited in some ways that again, poses a huge threat to, to humans.
So the risks are, are very large. And I think specifically to your question, I think, I think security companies could be some of the biggest winners from this rise of, of generative AI and there's going to be increasing need for, for products that sort of keep us safe.
[00:18:57] Sahil: Non technical solutions would be like developing a safe room with your family or for instance, you know, just common sense, but how could technologies be developed to protect people?
[00:19:05] Vinay: So there, there are all sorts of technologies that are sort of being developed now. One is just you know, you, you can obviously develop a model and train a model to, to emulate a human, but you can also like develop models that actually can detect. When it's an AI versus a human and be able to block the sort of AI detected content.
So like there are obviously like AI based approaches to doing this sort of stuff. I'm seeing a bunch of companies in the security simulation space, so developing like these simulations that help train employees and help them be aware of knowing like when they, they actually could be being, being smished versus, you know, when it's a real threat.
So, so I think there are a lot of opportunities. There are even things like just like fundamental new advanced and security that can that can help prevent some of these like prompt injection attacks and things like that. So the, the, the opportunities are pretty endless here. At the same time, I think like nothing beats some of the human prevention tactics of just I'm just figuring out a safe word and stuff like that.
[00:20:05] Sahil: Have you consider potentially shifting from the VC route and actually, once again, trying to be a founder and going to go build? If you would, like, where would it be?
[00:20:13] Vinay: Yeah. I mean, I think about it all the time. Absolutely. I think, I think this stuff is super exciting and this is like one of those game changing technologies that will really redefine the next.
You know, decade and certainly the next century. In terms of specific areas I'm excited about, I am very excited about what I call service as a software. So if you think about like the last decade of innovation, it's been driven by like the rise of cloud computing and what people call software as a service.
And part of my thesis is that like the biggest area where this technology is going to be applied is in like human services. So things like anytime humans are communicating with other people or like doing a professional service for another human. That's something where AI now has the capability to actually augment or automate that process.
And so now, you know, it's possible that you can actually run a services business, like a professional services business, but you augment or replace the labor with AI, and now you have a services businesses that run with the gross margins of a software business. And I'll give you a very practical example of that.
We at Foundation are investors in a company called Converse AI, which makes software that helps automate the phone screening process for staffing agencies. So, staffing agencies, you know, basically are like screening a bunch of people for jobs, and they have call centers often overseas that are actually like running these interviews.
And now it's possible you can light up this AI driven call center overnight that now suddenly Just interviews all your candidates for you. And so you can imagine like the staffing agency of the future has no employees is just an AI call center that now runs like a software company. And so when I talk about services, software, it's really this trend of like how do you think about services, professional services, businesses that are truly just like AI is in the background
[00:22:06] Sahil: to counter that.
Do you think there's any. Services, I guess, that can't be replaced by AI, call screening, that's one, but, you know, when you go deeper in like that same process, maybe not.
[00:22:16] Vinay: My, my view on this is, I think, it's all about the, the, how high stakes are these interactions, and how many, just, like, critical decisions are being made, right?
In, in the process of, like, phone screening a candidate, for example, like, you're basically helping filter the top of the funnel. If you take the other extreme example is like doctors interacting with humans where you're actually like making a diagnosis or something, right? Like, I think that's really scary and you want to keep humans in the loop as much as possible because those decisions are very high stakes, whereas something like in the healthcare context, you know, I'm seeing a bunch of really interesting companies that do things like.
Appointment scheduling, right? So typically when you call in to a doctor's office to schedule an appointment, maybe they verify your insurance eligibility, things like that. I think that's like a amazing use of generative AI today where it's like, you know, if the AI messes up somehow, like the stakes are, are actually fairly low.
But if the AI messes up, like, a diagnosis, like, that's, that's a big deal. And so I think initially that's how I think, that's sort of my framework for thinking about where it needs to be applied at first. But in the long, long term, I'm fairly bullish that, like, this technology is going to be good enough to actually slowly take over some of these really high stakes decisions.
But I think having a human in the loop is always going to be a really important component of this.
[00:23:41] Sahil: Last question I ask everyone I interview is, has there been any routine that you've kind of kept throughout the years?
[00:23:46] Vinay: You know, I think a lot about routines, especially it's the new year now, so I'm thinking a lot about resolutions routine.
I'm not a very routine oriented person. I want to be more of one. I always try to make time for my friends. Like that's something that's really really important to me is like, Every year, just making time to be there in person, like whether it's doing a trip or two or, or, or at least just like spending a few days in person with the people who are very, very close to me, that's something that like always just keeps me very grounded and very supported.
And so, yeah, I think it's like, it's just like spending time with those people and they also happen to be the people who like inspire me the most and energize me the most. And sort of help me keep my fire for what I do.
[00:24:28] Sahil: But with that, this has been UNinvested, I'm Sahil,
[00:24:31] Vinay: and I'm Vinay.
