Hello and welcome to another episode of This Month in Datadog. This month we have a very special episode for you. It's one I've been looking forward to for a while. Today I'm going to sit down with our co-founders Olivier and Alexi. Thank you both for joining us.
And we're live from New York City.
Exciting. It's exciting. So, one issue I had when preparing for this episode was there's far more than I want to ask than we have time for realistically. So, let's just get started and let's start at the beginning. Where did you first meet?
We met, we actually half met, right? We didn't fully meet. This was I think what, '96.
Yeah. Let's not reveal our ages too much, but you know, we were in college, not the same year.
And the story is interesting because I was part of the team that ran the campus network and Alexi was caught hacking into the campus network. So he was court-martialed by the team of students and then sentenced to disconnection from the internet and I carried out the sentence. So that was the extent of our relationship then.
Yeah. So he knew of me because I had done something, allegedly done something bad. But I was like, I just showed up, you know, in my dorm room it's like, oh there's no internet in my computer. What's wrong? And only much later did I realize this was a premeditated act.
It's definitely an interesting first meeting. And the second time you met, I realized it's a little cathartic for me to know that I wasn't the only one in a windowless room early in my career. And the windowless room I was in was also the server room for the dev servers. So it was a trifecta of no natural light, unbelievably cold and really, really loud. So when was the first time you actually worked together?
We worked together at IBM. So this was I guess upstate in Hawthorne.
Yeah. R&D. It was a windowless office. It was not physically cold but we were working on email protocols back then. So that's how when we actually formally met and worked together the first time.
Also it was not so cold because I don't know if you remember we blocked the AC vent.
So, do we have you to blame for POP3 or what?
Close. Close but not completely. Yes.
IMAP. Okay. Got it. So they say third time's a charm and viewers might be thinking this is where the Datadog story begins, but it's not. But to me, it is where the Datadog origin story really begins. So, can you tell us a little bit about Wireless Generation?
Yes. So, Wireless Generation is, I think it was the third startup we worked for or something like that. So we worked for a few startups. We worked together at the tail end of the dot-com boom and the full extent of the dot-com bust. So we've been through that which was very interesting in many ways.
Yes. In ways that are fairly familiar with what we see today. So we learned a number of things about what it looks like when you start a company. Also a number of things about what can go wrong and the mistakes you can make especially as you try and build new products to market. So we spent a few years there then we started working together at this education software company called Wireless Generation. And it's a startup that actually worked in the world of education which is a fantastically interesting space but also I would call the brutal space to build a business and build a company. And that's where the idea of Datadog was born. The backdrop is that Alexi ran the operations team and I ran the development team and we built the team pretty much from scratch. And we were very good friends, it was our third or fourth company together already. We hired everybody on our teams. We tried not to hire any jerks and yet the teams are fighting.
Yeah. And this was one of the early sort of real SaaS businesses because back then early 2000s everybody's starting to use Hotmail and so on but this was very B2C and so B2B as SaaS was fairly new so everybody was trying to figure out how to run it. And this was still, we were still operating as the dev team, you know, build the code, write it, it runs on my machine it means it's okay, throw it over the wall. The ops team is like, okay we have this, we have to run it in production, oh it doesn't work, it crashed, what do we do? And a lot of finger-pointing naturally happened. So that gave us early inspiration for Datadog. And the inspiration was not to build systems management, monitoring, any of that, we didn't come from that world. The inspiration was let's bring Dev and Ops together, let's get the teams to talk to each other. Let's not have the dev team coming to complain to me that the ops people don't know what they're doing or the ops team complaining to Alexi that the devs are cowboys who ship anything and they don't understand the applications.
And they don't call for it at least.
Exactly. Yeah. They don't wake up in the middle of the night to fix it.
Those were the days I guess.
Yes. And so the initial vision was we bring them into one platform, have them speak the same language and work together. It was also at the time when DevOps was nascent and in retrospect the thing we didn't fully understand then was that cloud was going to explode obviously and at the heart of the cloud explosion there would be a movement to bring together Dev and Ops much closer than it used to be, some disintermediation of the work of provisioning and managing hardware and all of that stuff. And that what we're doing, which is bring both Dev and Ops under the same platform, was actually going to be central to cloud adoption. So I would like to say it was all a beautiful plan that worked out. Turns out I think we're also pretty lucky in that we didn't fully understand the implications of what we were doing at the time.
I wish I could say it was a master plan though. I looked at the slides we put together at the very beginning of the company and we haven't veered too much. I mean we'll build much more than we had planned but we didn't do some 180 or something like that. So I think the way we thought it could work out ended up not being too far from what happened.
Interesting. So one lesson you said you learned there was we believe in learning over knowing. Can you expand on that?
So when you start a company I think there are two ways you can go about it. Two extremes. There's one extreme where you say we need to build everything from scratch, absolutely everything. And the other extreme is more you cargo cult everything, it's like oh this big company that's successful, they do it this way thus we should do it the exact same way. I think we try to find kind of a happy medium but always going back to what we can see, what the reality is telling us. I think we tried to be very pragmatic but not sufficiently grounded in what we had learned early on in Wireless Generation, the way you run teams, the way you think about product and stuff like that. So I think that we found our way effectively, that's my interpretation of learning over knowing.
And on the product side, again if you look back at our influences, we spent eight years building and shipping software and SaaS that was used by teachers. And the one thing we realized at the time was that we're not a good proxy for the customer. We have to learn from the customer and the user what actually matters, what we need to build, what's valuable. And we have to be very humble in that conversation that our ideas maybe they're interesting but really what matters the most is to understand the problem and what we can do and who we can make life better for these customers. So we didn't have that, like most of the folks who go into building dev tooling or ops tooling or security or all the things we do start from their own problems. We started from an experience that was building for others. So when we started Datadog the biggest fear was to build something that would not solve a real problem, a really valuable problem. And so we started from day one with that focus on the customer, the user, what their life was, what their problems were and how we could solve those problems.
So that brings us to Datadog. You mentioned that Datadog was not founded to solve the monitoring problem. What problem were you really trying to solve? What were you building?
