Jake Aaron Villarreal: I'm Jake Aaron Villarreal, born and raised in Silicon Valley. Here to take you behind the scenes to share what it's like to be a startup founder, the journey they're on, the problems they face, the products they build in an effort to make our lives better. I'm excited to have with us today Dima Syrotkin, founder and CEO of Pandatron. Dima, welcome to the show.
Dima Syrotkin: Yeah, it's an honor to be here. Thanks for having me.
Jake Aaron Villarreal: Well, thanks for joining. A little bit more about Dima: He is the leader of Pandatron, partnering with Fortune 500 companies like Panasonic, Mitsubishi, and KPMG to accelerate AI adoption. Originally from the Ukraine, he's based in San Francisco, and his mission is to enhance self-awareness and emotional intelligence globally. So what is Pandatron? It's an AI coach designed to help organizations navigate change with clarity, confidence, and care. So today we'll be talking about what the company provides, the problem it solves, and also some of the trials and tribulations as you're running a startup. Before we jump in here, Dima, where are you joining us from today?
Dima Syrotkin: I'm in San Francisco, yeah.
Jake Aaron Villarreal: Really cool. Well, I used to live there myself in the Marina, and lots of fun, lots of culture, diversity, and innovation. So...
Dima Syrotkin: I'm not far from, not, not too far from the Marina.
Jake Aaron Villarreal: Okay. Yeah. Very nice. God, what a view. I love that area. Before we jump into your company and what you're building and who it's helping. Uh, give us a little background of yourself. What were some of the experiences that shaped you as you grew up to eventually become an entrepreneur and get into technology?
Dima Syrotkin: Yeah. I mean, I think as a child, like one of the formative things has been, my grandfather was the CEO of like a local... it was a small town, 50,000 people, but...
[Dima's Journey to Entrepreneurship]
...it was built around like nuclear, and, and uh power like gas, oil/gas plant. And my, my grandfather was the CEO of the oil and gas plant, and it's kind of like the city was built around it. So he was almost like the, you know, the mayor almost, you know. And when I was growing up, like, I think he was already like ex-CEO, but like many people kind of like looked up to him. And that, in connection with me like having good grades, which was like the main mark, you know, when you're in school of like how people measure you. Everyone was like, "Oh my God, this guy," like I was like "what." So that, that really built my confidence being like a big fish in a very tiny pond.
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: That's... then, you know, going into, into the real world and moving to Finland at the age of like 18, starting studies, started working for this uh non-profit called AIESEC, which is like a student-run NGO, and um pretty early on there I had some experiences of falling flat on my face uh unexpectedly, despite having only success in life before. And um luckily I attended a conference on um, I mean fundamentally was focused on self-awareness and kind of emotional intelligence and those kind of topics. And I was like, "Wow, this is super powerful. Like, I feel like I was sort of swimming with the current before that and then now I finally kind of start questioning, like, what do I want, my fears, what are my dreams?"
And so in many ways that kind of changed my life and sort of set me up on this career that I've been on for the past um you know 12 years, where on one hand I, I kept the confidence, and that's why I became the startup founder I guess, that's overconfidence contributed to that. And then on the other hand, it was um kind of being mission-oriented and being like, "Wow, like I think this is super powerful," and then over time finding the niche where I think you can actually build a super profitable business on, on this. So yeah, that's um, those are some, some of the formative experiences.
Jake Aaron Villarreal: Really cool. Um, and what was the connection to technology? How did you get into that space?
Dima Syrotkin: I was good at math in school, and, but not good at physics. And so I was just like, "Okay, what areas like, you know, where I'm not going to be poor, but at the same time, you know, they have math, but they don't have physics?" So it was either applied math or computer science. And so I went to computer science. And so kind of like being in this student organization that I mentioned, very early on, I was like, "Hmm, I like this leadership thing, like getting people together and like project management and all of that, creating impact. This is amazing." Uh, but at the same time, I already like started studying computer science. So this kind of just merging a couple of my um, sort of uh backgrounds from very early on.
I realized that like, "Okay, I could become like a trainer," right, but I, I felt like doing things at scale is way more exciting for me. And, and I was like, "Yeah, technology is the path." You know, if I would just become like a coach or a trainer myself, there's only so many people I can affect.
Jake Aaron Villarreal: Yeah, I think that's a challenge for most coaches, is you, it's really your time and then, you know, your clients. And then, you know, from there, how much, how many can you put on the schedule for the week or the day or the quarter? And at some point you start looking at, you know, adding maybe people underneath you, and you can scale, you know, basically like a services business, but you know, technology is changing everything, and certainly in, in, in, in technology and in coaching and things like that. But walk, walk me through your company, so Pander... Pandatron. What, what was it that you saw in the market that you felt like, "You know what? There's a problem out there that hasn't been solved that I think we can solve," and how do you, how are you going about doing it?
Dima Syrotkin: So the first part of it has been working in management consulting, and that's where I was like, "Oh, like this, companies are really struggling with implementing the strategy that the consultants come up with," uh, which is called change management. Kind of like, how do you manage change? How do you implement, execute on the strategy? Um, and then at the same time, I was also relatively good...
[The Birth of Pandatron and Its Mission]
...at studying, like, somehow it's easy for me to get grades. So like after master's they were like, "Hey, we have a PhD's path, do you want it?" And I was like, "Sounds maybe interesting." So I like kind of like stumbled into a PhD and I started doing a PhD on the topic as well. So that only deepened my understanding of this. And I was like, "Okay, this is really like, this is a deep problem. Like people have thought about this for like decades. This is not..." and, and how to organize best. And it's almost like uh, you know, middle management is like the best worst solution. It's just like, it's not working super well. Like big companies are really stagnating, but we haven't come up with anything better. Everything else uh you know, has failed even, even worse.
And, and so then over time we were like, "You know, is there a technological solution here?" And, and hm, it might be actually connected to my passion with coaching, because if you were to create this kind of like middle manager or AI management consultant, you probably want to start as it being more of a coach um that, rather than like imposes something on people, like supports them, you know, um asks them questions. Um, and then on the other hand, you realize, "Okay, it's not really coaching anymore or consulting, because suddenly you can talk to, you know, a hundred thousand people at scale and collect a ton of data from those conversations, anonymize them, and then report that to the C-level." So those are like the two fundamental things we do right now.
On one hand, we have an AI agent that talks to thousands of employees, helps them understand the company strategy, helps them set goals that are aligned with it, helps them potentially even talk through the anxieties. Because no one was afraid of Workday, but people are afraid of AI, for example. So AI adoption is, you know, one of, one of our core topics. And on the other hand, we, we anonymize that data and then produce the reports for C-level to have oversight to understand what the progress is, what the systemic issues are, and what can they do.
Jake Aaron Villarreal: Yeah, you know, I worked for a big company, Oracle, and there was always changes that were things we had to work through or new applications they wanted us to use. And it was always this question of, is this helping or hurting me in terms of my own production? You know, I was a sales executive and you know, all of a sudden they created this, long time ago, but CRM. And it was a system that was not very well put together at the time. And you know, it was mandated to use it and put all your customer data in it. You know, ultimately the company would be able to see the, the forecast and be able to be more accurate. But getting people to buy into it and use it was a totally different challenge. So I understand.
Dima Syrotkin: Oracle, and Oracle is still a relatively modern company in comparison to something like, you know, I don't know, like the Panasonics and Mitsubishi of the world, and you know, Merck and whatever. Right. So tech companies are still having it easy because it's like, Larry Ellison is still alive and he's still kind of you know, so, so it's like, it gets even worse. You wouldn't believe it.
Jake Aaron Villarreal: Yeah, I can imagine. But you mentioned that 70% of major transformation projects fail to meet their timelines and budgets. We're talking about projects that you know, could balloon from 5 billion to 10 billion over 5 to 10 years. That sounds almost unbelievable. What's the most jaw-dropping example you've seen where this went catastrophically wrong?
Dima Syrotkin: Well, you mentioned ERP. So that was actually like, for example, Nike. They were implementing like an ERP, spent 400 million, 5 years. That was the plan. Sorry. That was the plan. 400 million, five years. Small project, ERP implementation.
Jake Aaron Villarreal: Wow.
Dima Syrotkin: The result, by the way, this is... this was just the plan. The result was more like... more like a billion and 10 years.
Jake Aaron Villarreal: Wow. That's incredible. And that's just implementing the system, right?
Dima Syrotkin: So that's um, it's, it's insane. Um, and you know, cloud transformations, like I, I saw that like, it was like, Amazon signed a contract with someone and that was...
[Challenges in Change Management]
...like 14 billion or something, cloud transformation. I think JP Morgan. Yes, JP Morgan. 14 billion. Out of 14 billion, a lot of it is, of course, going towards technology, but like, I think, like, was it like two billion of that was allocated purely towards training the employees? And training is a bit of a like a one-way street, you know? It's like you put a hundred people in the same room, you give them a lecture, and then you hope that they change their behavior and mindset after that. And, and often they don't.
Jake Aaron Villarreal: Yeah. Well, it's pretty interesting. You know, AI is being adopted or trying to be adopted in small companies and big companies. We were at an event a few months back. It was with a VC firm called GreatPoint in San Francisco. And there was, it was like a fireside chat, and there was um the founders actually, Ray Lane who used to be the president of Oracle, leads that company. He's the founder of that venture firm. And he was saying that out of all the, the Fortune 500 company CEOs, like everybody's talking about AI. What do they need to do to adopt it, to build it, to generate revenue from it, to reduce headcount from it? Like, the, there is this transformation happening across the US and also globally with AI adoption. But getting AI to actually get in the door and then to be worked on and used in proof of concepts, like what, what are you seeing from your perspective of AI adoption, and what are some of the things that are, are working and what do you think needs a lot of work to, to make it work well?
Dima Syrotkin: So I think like, fundamentally you want to uh you know, in the ideal world you would want to like unbundle the current like workflows. See like from first principles, like, what are we trying to do here? What's the whole process? And then kind of rebundle them with AI, because I think it's going to probably change a lot of how we do things. The issue, of course, is that like most companies, they need to keep up the continuity. And so you know, alternative approach, which I think could still work to some extent, is that you basically just like run small experiments and you replace small pieces at a time. And what we find, you know, what we help with specifically, is that number one, we help people understand what AI can and cannot do. Like many people view it a little bit like, you know, magic and they're just like, they have kind of like a wrong metaphor in their mind for like how to think about it and you know, what are the things it can actually do.
We also help them to identify the right tools for themselves and analyze their workflow and like, "Where are the opportunities here? What could we actually... where could we apply this leverage that, that we have, and what kind of tools, you know, could be useful? Are those tools already integrating? Could they integrate with your existing uh workflow, existing tools? And what is..."
[AI Adoption in Organizations]
"...your, you know, talking about again like this anxieties and emotional side of things, what is your kind of... okay, you might have certain like fears about AI. First of all, maybe like, let's make them explicit. Let's see if we can test them. Like, are they actually real or are they imaginary? Um, and maybe some of them are real. Um, and at the same time, like, okay, how can we mitigate them? But also what are the positives? You know, because what, what is the ideal like scenario here?" Um, because it's not always like all bad, right? Like there's probably a lot of opportunities here too that maybe you're not even, you know, paying attention to. Um, and then for the C-level, it's like, "Okay, let's aggregate what like thousands of employees are talking about. And then let's observe like, oh, like here this department, like even though you're, you know, it's very easy for you to see the front office, but actually the back office implemented something small but it's, it's producing huge results." Maybe you want to like study that, understand that, and then scale that across like other departments or other, you know. Because giant companies, they typically are like so big that it's like if something happens in one place, like people don't even know that, you know, in, in other places that that's working.
And maybe we, we then like tell them like, "Yeah, but you know your CFO office is like super concerned about your AI investments and you know, maybe you need to at least like triage that and see what's happening there." So kind of helping them to define a better AI strategy um of like what are the things to support, what are the things to kind of kill, what are the things to continue.
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: So that's, that's how we think about it. So it's, it's a, it's a big topic.
Jake Aaron Villarreal: Yeah. I mean 80% of business is psychology and 20%'s execution. So, it sounds like you're working a lot with the psychological side of things to get people opened up about, you know, the fears that are holding them back and opportunities that they can get on board with that are going to help the business. It's very, sounds like a real combination of psychology and business, which is really fascinating to me.
Dima Syrotkin: Yeah, I think both are important. And also the business side, right? It's like, okay, like we do need to still recommend the tools. Like we also need practical knowledge, too.
Jake Aaron Villarreal: Yeah. Well, one thing you'd mentioned in our previous call was that companies look at workflow automation and say, "That's AI," and really it maybe just makes you optimize how you operate your business. But AI is, is different. It's actually data that's, you know, can help you tell a different story or maybe give you brainstorming ideas about a product, or it's got some other advantages. When you work with companies, what's your, what's working best for you to kind of get them sort of, I guess, opened up about the reality of: Do they actually need AI and this billion dollar investment, or is it more about just optimating their people... optimizing their, their processes and the people to adopt those new processes to execute and to run a better company?
Dima Syrotkin: Yeah. Yeah. So in many cases, we started partnering with consulting firms because we realized that like that layer of like defining the strategy, talking to executives is also super important, but it almost feels like, you know, a consultant would be better suited in that regard. And so on one hand, it's kind of a go-to-market for us, but on the other hand, you know, we partner with them also to, to help the client like figure out those types of questions as well. And I think that like, you know, we definitely encourage more risk taking uh, because I think... well I mean I generally think that it's warranted basically. And of course you can sort of uh cap that risk, but I think like you probably want to think from the first principles like, "What is this going to look like in 10, 20, 30 years and what does that mean for today? Like, and how do we, how do we prepare? How do we uh sort of do that?"
And I think like, if we can inspire them with, with that, and... but also of course like outline how do we do this in a kind of, how do we de-risk it as much as possible, right? When I say you know doing something more ambitious, it doesn't of course mean that you're going to fail. Like, we recommend it because we think you will succeed actually. Um and, and I almost feel like sometimes like things that we consider risky... like people often think that like being a found-, startup founder is like a very risky choice for your career. I think...
[The Intersection of Psychology and Business]
...it's the opposite. Even if you fail, you're more employable than almost anyone else on the market. Like people really like appreciate the skills you, you learn. So I'm like, really the risk is so minimal in comparison to like, like I have, I've never like... I barely ever heard of an unemployed startup founder who like used to be a startup founder and then they couldn't find a job. Like, no.
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: And so I think similarly with AI, it's like, it's like you probably want to go all-in and think from first principles. Um, but then of course like, companies think about it differently. And for us the key was also partnering with like, like-minded partners, consulting firms that see the same future as us. And, and we, we kind of saying, you know, um companies, if you look at the stock market right now, it's like the tech companies are booming, everyone else is pretty stagnant. And um, there's not a lot of innovation. And once the company gets to a certain size, which is like around about half a million I've seen, there's like 20 companies in the world, it doesn't grow bigger. I feel like it's just like, it, it just kind of reaches that like plateau where it's like, "Okay, we, we reach a point of diminishing returns. Adding more people here will only make us less productive."
But I don't think it has to be like that. And I, I wonder you know, if this goes in the right direction, would this mean that kind of like, we will have you know most of the middle management work of like aggregating information, passing it along, contextualizing it, would that be done by AI? Right? And would we rather have like super empowered individual contributors who will have like, do the work of their lives and do like super creative stuff, and then like the C-level that actually feels empowered to you know do the best strategic work of their lives because they suddenly have the information and there's no like broken telephone both ways? Like right now where it's like, okay, there's 10 layers of hierarchy, by the time it reaches C-level it's super diluted, and then other way around, by the time it reaches front lines it's also like totally different message.
Jake Aaron Villarreal: Yeah. In some ways, it sounds like your offering or your technology is looking to optimize how a company operates, like taking a big enterprise company and almost operating like a startup or a small company where you don't have that middle management layer. You have the leadership with the direction and there's a strategy and let's execute, and you got the individual contributors doing the work. So is kind of like the vision to take the middle management out of the business and operate more...
Dima Syrotkin: At least augment, right. I don't know what it will be. Would it be replacement? But it's, it's easy to say, "Oh, we're going to replace things." But like, the reality is probably a bit more gray than that. Most likely it's not that you're going to like just like, you know, it's just more like that you know, some of them will transition to a bit more strategic roles that are a little bit more focused on strategy, and some of them will be focusing on their core uh sort of um, you know, core expertise. And maybe some others will be still monitoring the AI agents that are doing the kind of like the intermediation work, you know, um and do the oversight. But yeah, exactly. That's the vision.
But at the same time, you know, we can't just like... it's a tough pitch, right? Just be, come in and be like, "Hey, like we're going to like reinvent how you run your business," right? Especially for like giant corporations that are like you know, turning billions in turnover, and us being a startup. So then for us the question was like, "Okay, what's one problem that we can like start with?" And that's like where we found AI adoption. And then after that I think once we're successful there we would probably want to expand to like all kinds of different strategy execution, change management cases. We've worked a little bit with like, with innovation culture. So there's like a lot of different cases, cultural transformations that you can do.
And then the interesting thing is, once you have that data, can you then expand to like strategy creation as well, because suddenly you understand how the strategy gets executed. You get data from the front lines. Can you scrape the market data, understand what's happening externally? Can you get like internal ERP system data, whatever, and then you know provide the strategy to the organization as well, or at least again augment how the strategy is created? And then I think at that point, it's like you're basically augmenting the, the middle, middle management because on one hand you're like supporting the employees and on the other hand you are supporting the C-level, which is the role.
Jake Aaron Villarreal: Yeah. Gotcha. You know, there's a ton of proof of concepts out there today for AI and looking at carving out like one use case within a big enterprise and seeing if it actually works, and then if it does then you roll it out. And in that case, it...
[Optimizing Company Operations with AI]
...is AI adoption if it works because then you have to get people to use it, to build it, to scale it, to roll it out. Where do you fit in that space? Are you part of those proof of concept projects or you, do you come in at a different level?
Dima Syrotkin: Yeah, it, it depends on the, the, how the relationship starts. Very often we start with like a smaller pilot. And it's like maybe a couple hundred employees that are kind of involved and then you know the goal is to scale further. And then now we are negotiating like one deal where the client is like...
[AI Adoption as a Starting Point]
..."Hey, we're going to scale it right away to 300,000 people." But it also came not from the client directly, the deal came through a reseller that is like a giant consulting firm, that where like the CEO is involved talking directly to the this company, the potential client. So because of that relationship, it might go straight to scale.
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: So, so that, that happens too. Um,
Jake Aaron Villarreal: What's the business model?
Dima Syrotkin: Currently, it is like a simple per-seat. But now that we're negotiating this giant deal, we're being like, "Should we kind of switch towards cost plus margin?" Because the LLM cost is kind of linear and, and we don't want to like take kind of full responsibility for adoption. We want, pe-, the customer like prepay something that you know, they make a commitment so they onboard the people well. And on the other hand, we also understand the customer who is like, "Okay, like this LLM cost is quite high. What if we pay for it, but then people don't use it?" So instead, what you could do is kind of like cost plus margin, where you charge for LLM costs like separately and then you charge for the service separately. Uh so we're thinking about that.
And long term the goal is actually really switching towards a Palantir model, where you charge for results. And you know um kind of, let's say we would charge like 5% of like EBITDA that we like save and, and sort of create. Because if you assume that like, you know, like the goal would be to, to arrive to use cases that are very measurable, especially with something like optimizing the kind of coordination and the, the middle management layer, it's like relatively, you know, it, it's doable to measure what the savings are. And, and then can we actually make it... because I think that will make the move towards like going all in faster, because we can say, "Hey, look, you know, you don't, we're not like charging you per seat and just like trying to reach off this. We're charging you per result," right?
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: And we then, we are then staying, we are then becoming your like new middle management layer, but you know, you're only paying a fraction of what you used to pay. So you, you win anyway. Like if it doesn't work well, you didn't lose much. And if, if it works, then you, you saved a ton of money and made a lot of money on top.
Jake Aaron Villarreal: Yeah, I like that. Not reinventing the wheel, but looking at what someone else has done successfully, and you could, you know, copy parts of that and make it work with your own flavor. Palantir is a great company. We followed them for a long time. And what's interesting, we're in the hiring side of the business. So we help companies find people and grow and scale. The secret sauce that we're seeing from the hundreds of startups in AI that we're working with today isn't necessarily building AI that helps companies. It's making sure that it actually provides the outcomes that they want, which you're talking about. So, if they're going to invest in anything AI, they want to know it's going to make their company better and what outcomes are those going to be. In order to make those outcomes actually work, what we're seeing and the, the demand we're getting the most from the kind of the secret sauce of the successful AI startups are, are the role of the forward-deployed engineer.
Dima Syrotkin: Yeah.
Jake Aaron Villarreal: And this, for many that don't understand really what that is. Palantir helped create that role, which is really, it bridges the gap between making your product work well, vetting with the customer to make sure that they're onboarding the solution the right way, and then capturing what the customer says that they want, so you can take that information back to the engineering team and you could build those features into your product to make it more successful. And then you could take that and, and take that to another client and you've got a better product, a better platform. So we're seeing a lot of demand from AI companies for forward-deployed engineers. Is that something that you guys also have in your team and something that you are deploying when you're out there pitching your services?
Dima Syrotkin: 100% to the, to the extent possible. In our case, most of those clients, right? Like you really need to build credibility to get there because like you know, Panasonic is like you know just even simply because of like their security policies they're not going to just like let you hang out their office.
Jake Aaron Villarreal: Yeah, but...
Dima Syrotkin: But it's possible, but it, it takes trust to kind of, to get there. Ironically, our product actually helps with, with that, right? Our product is kind of a forward-deployed AI agent that then kind of like tells you kind of what, what works and what doesn't and kind of like brings that information back and brings that feedback back by talking to C-suite. So, this is exactly what we do, just in a more automated way, right? That's, that's also the way to understand what the product does. But we also for sure try to do it in a sense of like... I mean I think it's honestly like, it, it's, it also kind of became like a catchy word, but of course like you know since the dawn of time like for sure like you want to understand the customer needs as much as possible, and so the deeper relationship you have, the better. Um, 100%.
Jake Aaron Villarreal: Yeah. I mean we ask our clients all the time, so is this somebody that's going to go on site with the company and be sitting in a cube next to like the engineers, or you know is it async? Can you do this through Slack and email and chat? And it's about 50/50.
Dima Syrotkin: Possible. Yeah. Yeah. Also possible. But I mean, to me it's, it's more about the relationship, right? And the face-to-face helps with the relationship. That's I think the benefit.
Jake Aaron Villarreal: Kind of need that.
Dima Syrotkin: But for sure you can, you can definitely do it when... with most clients we, we kind of forced to at least for now do it remotely for sure.
Jake Aaron Villarreal: Right.
Dima Syrotkin: But that's um, in a sense it was like the role of the account manager before which it was now like adding a little bit and I think like adding tech people in the mix like again has always been there, but now it's making it like a little bit more explicit that like, "Hey, it's not just like a salesperson. This is like an engineer that can understand your needs and like kind of almost code them on the fly." It's just like, you know, um, you know, the, the time kind of like decreasing the, the sort of the, the, the time. But yeah, um, in many ways like a lot of uh, new things are forgotten old things.
Jake Aaron Villarreal: Yeah. Well, I think you're onto something with what you're building, and I'm excited to see where it goes. I want to talk a little bit more about the AI coaching experience. Walk me through what it actually feels like to be one of those 100,000 employees talking to your AI agent. I'm a middle manager at Panasonic. I'm confused about the new strategy. I'm resistant to the change. What happens when I interact with your system?
Dima Syrotkin: So, we initially do a kickoff for the people just to give them an introduction and be like, "Hey, here's what we've done before. Here's all the fantastic feedback from different people." We give them like a five minutes kind of experience where it's just like a very quick chat that helps them with like understanding what their top goals are, helps them reflect a little bit like "What's my biggest challenge today, whatever," and kind of does a little bit of like coaching of like, "Okay, what's the next step?" Right? Not so much pushing an advice on them, but like more like, "Okay, let's help you like what's the plan? Um, and, and let's, let's like note that down. This is, you know, a good, good kind of point in the end."
And then we integrate with the calendar. So, we, we ask them to right away be like, "Hey, like book a recurring slot in your calendar." So, it's typically like a weekly half an hour. And then off they go. And they end up kind of going online. It's a web app. The core functionality is effectively via chatbot. They can, you know, talk to it, you know, via voice. A lot of people do. They can type if they're in a public place. And it's, it's kind of like, you know, talking to, to a very like smart expert who is not like your manager. So you like, on, on one hand, you're like safe because like they're not going to fire you. They're not going to judge you. It's just AI. But on the other hand, like they know your company strategy. And rather than like pushing advice on you right away, it, it like helps you understand as well like what do you believe in, what do you stand... helps you kind of set goals, follow up on them, uh talk through those things.
Every couple weeks it will send you like a summary of what you've, what have been discussed. There's some gamification elements, like there's a streak and you can see like how many, how many kind of weeks in a row you've, you've went. And uh there's a summary of the previous session that kind of when you come online you're like, "Oh, this is what I discussed last time. Interesting, my goals kind of changed." And you know, and it can be both like I've, I remember I was like talking about like anxiety and like very like emotional things. And then the next day I was like, "Actually, like I'm curious if you can recommend me any AI tools for this and this purpose." And um, a lot of it is like this contained 15, 20 minute exercises that are built by experts. The challenge with ChatGPT is that it's so broad that you kind of like really have to like know what you're looking for to prompt it in the right way. Um, and we basically create this like very contained exercises. So, for example, if you're looking for an AI tool, we have like a module just for that, right? That also knows your context of the organization.
[The AI Coaching Experience]
It knows background about you. And uh, and it's like, "Yeah, like you know, if you're looking for something super, you know, reliable, CodeQL is actually owned by Microsoft and you are using GitHub already, so this is like as reliable as it gets, this is you know as, as close as it is." Um, and then we also... so I mentioned this reporting we do for C-level. And in the beginning we were afraid like, will people be skeptical about like the Big Brother vibes there? But they actually see it as a way to express themselves because they say, "Okay, like this is my like main option for like where could I point certain issues and not be sort of um afraid of being like singled out or something." And, and if that issue repeats with my colleagues, then I know it will be escalated and, and then like the, the people that need to see it will actually see it. Um so, so they sort of see it as a way to, to sort of express themselves. And then yeah, we do some like monthly like user meetups where they can like come and ask questions and share their experiences. So that's a little bit about the experience.
Jake Aaron Villarreal: Yeah. Yeah. Very cool. I love that you partner with some thought leaders on psychological safety and organizational power. Give me one framework or mental model from your AI agent that someone listening could use tomorrow in their own team, even if they have like, you know, 20 people.
Dima Syrotkin: I mean the biggest like, and this is maybe something you already used uh, but fundamental thing that comes to mind... like I'm a busy person. Priorities are super important, right? It's, it's always the core I think of any work, is like what are your priorities, what are the right things to focus on? And there's a few frameworks there. Like one of the very famous one is like, what are things that are urgent and important? And that would be like one quadrant. And those are the things you should do now, right? Because they're urgent and important.
Then, okay, what are the things that are important but not urgent? Those are the things that you end up postponing constantly because they're not urgent, but they actually are important, so you should get them done. So like with those ones you really need to put them in the calendar and like make sure that you do, do them. And if, if it doesn't happen, you need to kind of follow up and see like... and in our case, the AI agent can follow up on that and be like, "Hey, like, have you actually done that important thing you want to do?"
And then there's things that are urgent but not important. Those ones ideally delegate, right? Because they're urgent, someone has to do them, but they're not important. So maybe you're not even the right person to do them if they're not important because you want to focus on important stuff. And then the last bucket is like, not urgent and not important. Then well, just like, don't do that stuff. It's like, that's probably the stuff you can keep postponing for now and then at some point maybe you just cross it out from your to-do list of, "Actually, like, maybe we don't even need to do this if I look at it very honestly."
So this is like a very simple framework, right? And the difference from something like, you know, ChatGPT is that ChatGPT tries to like give you an answer right away and it's just like kind of like Google on steroids kind of like, "This is what you have to do." With our tool, it's more like looking at the long-term usage. And it's like, "Okay, so let me ask you a few questions. Let me help you kind of bring your problem into this framework and ask you questions one by one. And then let me follow up long-term and see how that situation evolved and, and maybe what, what do you need to learn from it? What, how do you need to change your behavior in the future?"
Jake Aaron Villarreal: Yeah, well said. I like that. Let's think 2035, the vision going forward, and your company and your technology has worked at scale. What does the business world actually look like? Are there fewer failed transformations? Are there massive companies actually innovating faster? Paint me the picture.
Dima Syrotkin: One of the interesting outcomes I think is that while we talk a lot about automation and companies becoming leaner, I think if what we do works, some companies could actually become bigger because suddenly there's no ceiling on a big company being effective. You can actually create bigger impact and you know people in the company, when you ask them like, "What do you do?" They're not going to be like, "Oh, I work in a big company," and there's this like sigh of like... and everyone understands that like, "Yeah, big companies suck, so like, yeah, I'm sorry for you." Like, "You're probably just like a cog in the machine and you like feel like it's horrible." But like, that would change, where suddenly you feel like, "No, actually like I can do like the best creative work of my, of my life, you know?" Um, and, and suddenly you...
[Frameworks for Team Effectiveness]
...know, I think the vision is that maybe those big old companies can reinvent themselves. And you don't need a crazy visionary person coming and dictating things top-down, because first of all, the C-level has more resources. They are more capable of making better decisions. And second of all, everyone on the front lines is also more empowered. They have more agency. Suddenly everyone is a little bit more like, "Hey, like what if we do this, and this sounds like a great idea." And it's like, "Let's, let's scale this innovation." And so that you know, we don't need the sort of... nothing against Elon Musk, but sort of that... it's not going to be a requirement anymore to have someone insane to come in to kind of reinvent an old industry like you know, the space industry. Sort of, NASA could actually innovate themselves.
Jake Aaron Villarreal: Yeah.
Dima Syrotkin: Rather than like waiting for the savior to come in and like do the work for them externally with some, you know, crazy visionary founder. Um, and, and again, I... nothing against visionary founders. I think they will still be there and it's, it's, it's a treasure, right? Uh, but still kind of like having a bit more bottom-up innovation. I think people will be... if this truly works, I think these tools will be also available for consumers on just more like a personal level, and you would be able to really flourish as like, as a human being. I think you know whatever, people will be 3% more fulfilled. And that 3% could really accumulate if you look at like the global scale, where suddenly, you know, people ask themselves, "What do I really want? Like, what does a good life look like for me?" And then they dare to take those steps. They find that courage with that bit of external support.
And, and I don't think, by the way... like I think part of that kind of better life would also be relationships. 'Cause some people are afraid of like, "Will AI replace relationships?" I don't think so. I think it's, if anything, it could help us to be better in relationships uh, and kind of you know, support us there as well. And then the crazy idea is like, you know, you know, somewhere in 2034 we start coming to the governments and be like, "Hey guys, like we have a lot of data on what the population truly needs like on a fundamental level. And could that inform policy? Could that inform what really..." because currently it's like you know, the politicians I feel like are you know, guessing. Do they really have data on like what, you know, what is needed in society? What are the different, you know, things, and, and what is the goal? Like the goal is GDP like for what? Like I mean, I mean kind of it makes sense, it's like, okay, the richer you are, the more resources you have like, you know... but, but can we also, you know, pay a little bit more attention towards like um, how fulfilled people are? And you know, whether they actually, you know, enjoy their lives and whether they're doing something that's ethical. So this is some of the ideas of how the future...
Jake Aaron Villarreal: Yeah. Well, I love that future. It looks...
[Envisioning the Future of Business]
...bright. You're very pro-AI, so am I, so is our company. And we've built agents ourselves, and we've looked at supporting companies in different ways with our technology, and it's been fun to see that evolve, too. And as we look into 2026, which is just around the corner, what are you excited about? What's on the roadmap for Pandatron?
Dima Syrotkin: For Pandatron, so one of the things I'm excited about is currently we have like this individual level layer, right, where we support individuals. And then we have this like organizational level layer where we provide intelligence to the C-level. What's missing currently is the team level, the, the group coaching. So one test I did is I asked people like, if, you know, at the end of the session maybe you set a goal, and then there could be a pop-up saying, "Hey, you know, there's someone else in your organization working on the same problem. Would you love to connect with them and exchange ideas?" Um, people, 85% of the focus group said, "Yes. I'll, I'll click. Let's try."
And when I dig deeper, they said like, "Oh yeah, you know, in this giant organization, I never know what other people do. And like um, I am you know honestly like sometimes even kind of feel lonely even though I'm like working with like technically a lot of colleagues. But like, I don't know anyone. And so if we could like work together on certain things that are like relevant, that, that's, that's exciting." And, and in comparison, like that was very different from some other features. 'Cause I also asked for example, "If you get an insight in the session, would you want to share it on LinkedIn?" And like only 5% there said yes. So 85 in comparison, this feature was like blew out of the water. And I was like, "Wow, this is very like popular, really, really want that." So, and I think it really kind of you know, makes it more holistic as an approach.
And then the other thing we're working on is a lot of integrations. So just making it like understand as much relevant information as possible. And maybe, you know, like, we're just one tool. We're not all, you know, be-all. So kind of like, "Okay, let's say, you know, could we connect you with uh some course where you realize like, 'I want to improve in this area.' Could we then like pass you to that e-learning?" Because one like angel investor uh said like that it's not the materials to learn that are scarce. It's the motivation. And I think no one is like really using e-learning because it's often like not timely and it's like hard to find the right thing. But if you could, like the coaching and that consulting aspect could help you find like, "This is organizational strategy. This is how it relates to me. Here is a gap," and you like set a plan. You're like, "Okay, this is my goal." And then it's like, "Okay, the last mile in it is like, 'Yeah, here's the material you need.'" And then it kind of fits in the bigger picture. Then you're like, "Okay, this is why I'm learning this thing. Like, now it makes sense."
Jake Aaron Villarreal: Yeah, I think I mean, I can totally relate to that. You know, working at big companies prior, you always sit down with your managers, and this is really middle management, but you know, they give you guidance of what they think that you should do, you know, for the next quarter. And then, "Here's the courses education-wise you should look at taking to get better at your job." And like that's good, but it would be so much easier if you could just go into like your company AI, say, you know, "How am I doing with the quarter? Here's the things I'm trying to figure out and here's what I'm trying to learn. And I align with the company strategy and what we're trying to execute. And if it could tell you those answers, and by the way, coach you up on here's the things you should be also learning, and here's where you get those videos or those trainings or whatnot." I mean, I think like, at an enterprise company I think that would be very helpful. And even at smaller companies, it would be so. Yeah, I really, I love the innovation. I love the space you're in. I studied psychology before I got into business, so I'm very much about the human factor. And uh, it's all about you know trying to find the outcomes that make people operate better. You know, your, your technology is, is a combination of, in my opinion, like psychology and technology and making, making people buy into what, what...
[Innovations on the Roadmap for Pandatron]
...the future is. If anybody wants to find you, Dima, or wants to find Pandatron, where do they go?
Dima Syrotkin: pandatron.ai is the website, and then yep, there's contact information there. And then alternatively, I'm on LinkedIn, Dima Syrotkin. You can uh look me up there as well. Yeah, those are some of the best places.
Jake Aaron Villarreal: Yeah. Okay. Very cool. Well, thanks so much for coming on the show, and thanks for the listeners for listening. It means a lot to me. You spent your time with us today. I'm your host, Jake Aaron Villarreal, signing off for now, but can't wait to catch up with you all in the next episode. Until then, Dima, the world, take care. If you like what we're doing, don't forget to subscribe, leave a review on Apple Podcast or wherever you listen, and follow us on YouTube where we go behind the scenes to learn what it takes to be a startup founder.
.jpg)
.webp)
.webp)
.webp)
.webp)
