The GitLab way: Kindness, transparency, and short toes

Lenny Rachitsky
You guys share so many of the things that most people keep secret. You post videos of your team meetings on YouTube. We get people who then contribute because of what they see. Oh, I can go build that. I know what that is.

David DeSanto
Hey, I ran that same problem too. I'd love to hear how you solved it. You also have this handbook, how you onboard people, how count payables works. At GitLab. We'll have companies who will fork the handbook if you can leverage what GitLab has done.

That's amazing. You mentioned the value of short toes, which I was definitely gonna ask about. That's where if you have long toes, you feel like people are stepping on you. Whereas if you had short toes, the work, it's not about you. You end with a lot less of the negative headbutting, especially in an asynchronous culture like GitLab.

Lenny Rachitsky
Is there any tips you could share with PM's that are just struggling in a remote world? If everyone's really annoyed at you, you're probably actually doing your job well.

Today. My guest is David Desanto. David is the chief product officer at GitLab, which is an incredibly unique company. It's the largest remote only company in the world. They share many of their team meetings on YouTube and theyve grown from being just a source code management business competing with GitHub to a multi product platform that covers security compliance, continuous integration, project management, deploy tools and more, many of which are infused with AI magic.

In our conversation, we dig into Gitlabs culture of transparency, including how they operationalize it, what they share publicly versus what they dont, the surprising benefits of working this way, and why its worth considering going going transparent with your organization. Also explore some of GitLab's other unicultural values like kindness and having short toes. Plus, David shares a bunch of great advice for scaling remote work based on what's worked well at GitLab over the years. Also, when it makes sense to go breadth over depth versus depth over breadth when launching new product lines, and how it worked for GitLab over the years. This was a fascinating conversation.

And if you want to learn more about a company that's doing things very differently while also kicking a lot of ass, this episode is for you. If you enjoy this podcast, don't forget to subscribe and follow it in your favorite podcasting app or YouTube. It's the best way to avoid missing future episodes and it helps the podcast tremendously. With that, I bring you David DeSanto after a short word from our sponsors. This episode is brought to you by Orb.

As a business, you care about revenue, but as a product team, the last thing you want to do is delay a product launch or a pricing change because your team has to rebuild billing from scratch. Orb is a flexible, usage based billing engine that lets you evolve your pricing with ease. The fastest growing product teams at companies like Vercel and relet trust Orb to power their pricing changes and launches. Use Orb to ship product faster. Stop worrying about billing and evolve pricing with ease and control.

Check it out@withorb.com. Lenny and skip the line for a demo or sandbox by using promo code. Lenny that's withorb.com Lenny this episode is brought to you by EPO EPO is a next generation a b testing and feature management platform built by alums of Airbnb and Snowflake. For modern growth teams, companies like Twitch, Miro, ClickUp, and draftkings rely on EPO to power their experience experiments. Experimentation is increasingly essential for driving growth and for understanding the performance of new features, and EPO helps you increase experimentation velocity while unlocking rigorous, deep analysis in a way that no other commercial tool does.

When I was at Airbnb, one of the things that I loved most was our experimentation platform where I could set up experiments easily, troubleshoot issues, and analyze performance all on my own. EPO does all that and more with advanced statistical methods that can help you shave weeks off experiment time. An accessible UI for diving deeper into performance and out of the box reporting that helps you avoid annoying, prolonged analytic cycles. EPO also makes it easy for you to share experiment insights with your team, sparking new ideas for the A B testing flywheel. EPO powers experimentation across every use case, including product growth, machine learning, monetization, and email marketing.

Check out epo@getepo.com Lenny and ten x your experiment velocity. That's getepo.com Lenny David, thank you so much for being here. Welcome to the podcast. Thank you for having me. Very excited about our chat.

I'm very excited as well. You definitely win for best beard of any guest of the podcast so far. What does it take to maintain such a beard? That's a great question. I would tell you it's not as much as you think it would be, it's just regularly going to a barber you trust that can keep the shape.

David DeSanto
And it does get washed and conditioned regularly because my hair migrated south for the winter, never came back, and so it's kind of become the staple. So yeah, comb it, put conditioner in it, things like that. Not a lot of maintenance, surprisingly. Okay. And I think you mentioned offline your handle.

Lenny Rachitsky
Is that on Twitter? What is it? Oh. So I just started trying to use threads and threads. Okay.

David DeSanto
Everywhere else, I'm like D Desanto or David Desanto. Those weren't available. So I'm David the beard. David the dot beard. Amazing.

So you can all follow me there once you watch this, because, you know, I think I've got ten followers in the first couple of weeks here. I would love to get it to a much higher number. Yeah. So you can help with that. Okay, let's work in that.

Lenny Rachitsky
And then, if you're not watching us on YouTube, you're missing this beer. Anyway, I'm excited to have you here because I've always been really fascinated about GitLab's culture. There are so many unique elements to how you guys operate, and clearly it's worked out really well. The company I was just looking up, the stock is a $11 billion business at this point. It's been around for a long time.

So I want to dive into a number of the different ways that you guys work, especially the stuff that's really unique. And the first element is your very unique culture of transparency. I have not seen anything like this elsewhere. So, first of all, as an example, you post videos of your team meetings on YouTube. That's correct, yeah.

Okay. And I want to chat about what? Like, the policy. But just as an example, I was watching a product team meeting of your product team a while ago where they're talking about a book club of Barney Kagan's recent book. I was watching engineers talking about some scaling challenges they were having.

There's a video of you being introduced to the UX team from a year or two ago, and maybe it was, like, after a reorg or something. What's just, like, the policy on which videos go online? Yeah. So that's a great question, and you're actually correct. So about two years ago, we moved Ux from engineering into product to make them a much more strategic part of the company.

David DeSanto
I knew a lot of ux, but didn't know everyone in Ux. So we took the time to let me introduce myself to them if they weren't familiar with me or hadn't talked to me at all. But to your question, our policy is like, be as transparent as possible. So if it's not customer data, it's not vulnerability information. We heavily encourage the team to put their videos, like, record their videos, sometimes even live stream them onto our YouTube channel.

So for those not familiar, you can search for GitLab unfiltered on YouTube, and there's more content there than a human being could watch in a single day, or even everything that's recorded during the day, because there could be multiple meetings happening at once that are also showing up there. But a good example of that is my bi weekly product meeting where we go over things, were working on things, were excited about troubles, were having challenges, and how we can work better together. I can tell you that it was a change joining GitLab in 2019 to have that much content just be available. You also learn how to be better on meetings because of it. My first meeting was my first day, and I spent a lot of time looking all over the room because I was used to being on a webex where you can't see anything.

So you could be looking outside while talking, and all of a sudden you realize you're not really engaging because you're assuming it's just an audio call. And so I think it's been really good for the company. The great example I can give you is we get people who then contribute because of what they see. And so we've had customers, open source community members, go, oh, I can go build that. I know what that is.

And they will just knock it out and do a code commit, or, you know, we'll get feedback on the video or feedback on social media. It's like, hey, I read that same problem, too. I'd love to hear how you solved it. And so I think it's been really great for not just the company, but the broader community, whether they're our paying customers or open source contributors. So just to understand there's developers watching team meetings, noticing there's like a bug or an issue or challenge, and then just going ahead and committing code to fix it.

Yeah. So sometimes it'll be open source. Yeah. Community contributors who are doing that. So, yeah, they'll hear about that.

They'll look at the issue tracker because the issue tracker is all public, and they'll be like, oh, I can, I can do that. Let me just do that and make the project better. Oh, my God, that's crazy. Okay. And so the policy, as you described is it's up to the person they decide, is this something we can share that we feel comfortable with?

Yeah, it's up to the person and the, the team. So, like, I do have a couple policies for the product division where we don't post the recording we used to do live are what we call our performance indicator reviews and key reviews. We stopped putting those public because we want to talk about the customer that's impacted or we want to talk about the thing that we need to move. And so we'll talk about it at a high level in our product division wide meeting that is posted. But then the nitty gritty down details that would be considered material, non public information or customer data that would need to be kept safe.

Those things, we will record those but keep those internal and not make them public. Okay, amazing. So you have these videos, you also have this handbook, the GitLab handbook which is handbook dot GitLab.com which also shares tons of pieces of how you all operate. So just skimming through it, there's your mission, your vision, your strategy, how you onboard people, your anti harassment policy, how count payables works at GitLab. Hilarious.

Lenny Rachitsky
Amazing stuff. Its great too. I dont remember how many pages it is now but its huge. And so what I see what happens and this kind of gets into the first example, well have companies who will fork the handbook because its just source available too. At the bottom of the page you can click view source.

David DeSanto
And I just heard from a company whos now becoming a customer who we were like we wanted to redo our UX mission and how we operate in a UX department. So we cloned the UX handbook within handbook dot get lab.com and it became our baseline for building what we do. And you can see that even for startups who will say theyve cloned the entire handbook and theyve used it as a way to start. And I think for me, across my career ive been in this situation where ive needed to fix something or restart something or, you know, I'm the first person in the department and there's a lot of uplifting you have to do in that role. Whereas if you can leverage what GitLab has done, that's amazing, right?

It really helps you continue forward really quickly. Yeah, I think one of the most interesting for me has been the competencies for product managers. Something a lot of PM teams are looking for is like how do we level and ladder and create levels for teams? And you guys share all that and you guys have a really good version of that publicly. So we talked about the YouTube.

Lenny Rachitsky
I have a bunch of questions about how this actually works. So you have the YouTube channel of team meetings, you have the handbook. Is there anything else that I'm not mentioning that's public that other companies don't share? I think there's really two things. So the first one is our issue tracker for everything we're doing, the majority of those issues are public.

David DeSanto
It's also available for people who have an account to also create and comment on. Whenever we do a call with a customer or I present at a conference or my team presents, we always say, hey, please go look at our issue tracker and if there's something you like, vote it up. If there's something you want to see change, leave a comment. And I think that's unique to GitLab. I think we've actually caused other organizations to do something similar because of that.

And the other one is our direction and our strategy mentioned earlier. Sometimes it's just a high level paragraph, but sometimes if you go deeper, we have like a one year direction that is very detailed and links over to our issue tracker to help you see like, hey, we're going to do, I don't know, we're going to build widget X and we're going to do it in the next six months and here's how we're going to do it. And I think those are, those, all those things together have really allowed GitLab to be the most transparent, publicly traded company in the world. And we're very proud of that because it's just in our DNA to do it. I think what's really interesting about this is it's kind of the epitome of it's not the idea, it's the execution.

Lenny Rachitsky
You guys share so many of the things that most people keep secret and it's worked, I guess. Is there any lessons there? Just like it's actually a lot. The execution is what separates you from everyone else. Someone could just basically copy everything you're doing and in theory build something like you built, but they haven't.

David DeSanto
Yeah, I think you're touching on actually my, when I interviewed in 2019, I actually asked Sid, our co founder and CEO, a similar question. I said hey. And I came from security into GitLab and I said hey, we dont share more than four or six weeks with a roadmap, with customers and even sales because we were in this situation where we didnt want our great idea to be taken and built before we built it. And that had happened to me at other places ive worked. And so in that conversation with him, he basically said products job to be ambitious and Zengineering's job to meet that ambition.

And if you think of it that way, and it's that collaboration between the two, it becomes less scary to put the information out. Uh, but you are right, we're very even transparent there. The thing that gets into what you mentioned the ability to ship software. I'm unfloored by these numbers, but these are real numbers for us. We, we still ship twelve releases a year.

It used to be on the 22nd of every month. It's now the third Thursday. So we don't have to have people work over a weekend. But weve done that for over a decade. And our last release put us at 149 releases that have come out essentially every month for the last little over twelve years.

Lenny Rachitsky
Someone listening to this may be like, okay, wow, we should try this sort of thing. Lets try to be transparent with everything, meetings everywhere, public handbook. I imagine this doesnt work for most companies. So a question just on my mind is what do you think it takes for a company to work this way? Like, what are some elements that need to be true for them to succeed being this transparent?

David DeSanto
Yeah, I think it is pushing yourself to realize what is actually confidential and what actually is not confidential at other places ive worked there end up with these artificial silos, even if youre all in the same office, and that ends up impacting your ability to truly collaborate and get the results you're looking for. And I can give you like a great example that sometimes you have a program manager or program management team who's tracking the schedule, and now they're being forced to be the router between teams, as opposed to say, it's all stored in a single source of truth. Anyone can comment on it, anyone can contribute to it. And that starts all the way from like team meetings, whether it's my leadership team or it goes all the way down to coffee chats that we have, where sometimes people just record that chat because all of a sudden there's something really interesting in that. And for those listening or watching us, as you said, looking at the beer that I have, instead of just listening to the podcast, I think they can.

Lenny Rachitsky
Hear the beer too. It gives me presence. So, yeah, makes your voice deeper. There you go. What ends up happening for us, a coffee chat is the equivalent of walking in and running someone in a kitchen.

David DeSanto
So you're getting more coffee or tea or water, right. And so sometimes those just end up getting recorded because there's something that's a good nugget of information that comes out of it. And so realistically, you just have to push yourself and it almost has to be uncomfortable. And then you realize you're starting to really, truly be transparent and you're allowing everyone to contribute. I imagine this isn't all upside and rosy and rainbows all the time.

Lenny Rachitsky
Is there anything that you've heard of or ran into where this caused a problem, either for the company or someone internally? Yeah, that's a great question. So I think if you go back to the like, you have to push yourself to your uncomfortable. Sometimes it ends up being overly transparent. And GitLab has had its fair share of things at times where an issue is public that probably shouldn't have been public, like in our issue tracker or, you know, the recording accidentally gets set to public when it should have been set to private.

David DeSanto
Uh, and those are just pain points and learning. And then as long as you're reinforcing that learning across the team, you end up not making that same mistake again. And I will say that I think the risk of that occasionally happening is way below the value of actually pushing yourself to do it. So what I would encourage you to do is start as simple as publishing a team meeting and making that available for everyone in the company. Begin to go from there to maybe like weekly meetings or even asynchronous readouts that you can post on Slack that are just saying like, hey, here's everything that happened this week for the team.

And then you'll slowly find that if you're doing it, others are doing it, and before you know it, you've actually achieved a high level of transparency. Now, depending on the industry you're in, maybe you can't make your issue tracker public. If you're like GitLab and you're using GitLab.com, maybe you can share the details we share about customer meetings and our conversations with them, obviously without saying their name, but you'll find the right balance for you. And I think you'll truly find that as a company you become more productive. To convince someone that this is worth doing.

Lenny Rachitsky
What do you find are the biggest benefits of operating this way? Because it sounds like, it sounds like more work and it sounds like there's more risk, what do you find are the benefits and also just maybe second order benefits that you didn't directly expect? Yeah, I would say the big thing I've seen is focus on results, especially for our customers. We are a global company. We weren't always as big as we are.

David DeSanto
But GitLab is over 2000 people. It allows people to asynchronously consume what happened while they were offline, gives them both more engagement as well as that lack of feeling of like FOMO or fear of missing out. And to me, those are really the top level items that come out of it. The secondary benefits of those are, of course, teams feel better aligned. You find things that you may not find till the end of a software release or beginning of a go to market campaign.

You have to redo something. And so I just think really, and I say this with encouragement to people, pick one project and try it. And if thats successful and doesnt feel like a lot of additional work, try it with another one. And before you know it, you might find out that its actually a lot easier to be transparent than it is to not be transparent. Because now you dont have to worry about tracking things you may or may not have said.

Think about, well, did this person hear it or were they not in that meeting? It just becomes thats a lot of data that people can consume and then they can make informed decisions quicker because they're not being siloed out of the conversation. Well, it feels like there's two elements here. There's being transparent internally with here you can listen to every meeting that you've had as an employee. But where you guys go is one step further, is anyone can pay attention to this and read your handbook, what do you find are the biggest benefits of that element of just like being public about this stuff?

Yeah, I mean, for me it really comes down to the engagement we can get externally because our issue tracker is public, our customers comment on it. We get community contributors committing to our code base, helping get be a better product. I can see where in certain industries we'll say, like financial services, maybe you can't do that as much, but I would say that there's still a balance where there's probably something you'd be more transparent about. I'm even seeing in heavily regulated industries where they're starting to publish more of the roadmap externally as a way to build more trust with their users and their customers because now they know where you're going and it's not a cloak and dagger type feeling. And so I will admit it can't be done everywhere, but I think it can be done a lot more than it's done today.

And you get a community feedback loop, which is great, especially when being at a product division. So one takeaway here is it feels like most of the benefit is in products that are building with a community component and developer oriented and open source. Feels like those are like the Venn diagrams of where this is most impactful versus slack doing this or, I don't know, Salesforce. Yeah, I mean, I would say that if you're in tech and you're building a product, regardless if you're a developer platform or devsecops platform like us, I think there is benefit. I think Salesforce could have a little bit more engagement from a broader customer base.

My last employer, we started sharing the list of open requests for new features more externally and we got a much more honed in roadmap and that led to better customer engagement which led to a higher retention rate, which led to better expansion numbers. Right. So I think its about whats right for your business and not necessarily are you in column a or column b? I was browsing around through the handbook and I looked at your core values and there's some really sweet things in there. So within the collaboration core value there's a value of kindness, which I love.

Lenny Rachitsky
What does that look like in practice? At GitLab the values really come together in a way that I've not seen other company values come together and I don't remember where these all are in there, but it ties into the kindness. If you're all remote, you're not necessarily seeing people in person. And sometimes it's hard to read intent on a slack message or the emotion in the message. And so we say assume positive intent.

David DeSanto
Assume the person is traditionally just asking for help or is trying to be helpful. Have short toes like don't read something and think it's meant a different way. And that's where the kindness comes in as well, is that if you're treating each other with kindness and assuming positive intent, you end up with a lot less of the negative headbutting you can have, especially in an asynchronous culture like GitLab, our team, I will see my direct reports, usually at least a couple minutes a week on a zoom call. But that's not the case for everyone at GitLab, right? And same thing with other teams.

And sometimes it's hard to assume what the person's intention actually is. And so we say be kind, assume the positive intent. Realize that everyone's here to help each other out. And I can tell you when people ask me like, what's it like working at GitLab, I always come back to the values and say we actually live them. And that starts with Sid all the way down the organization to a brand new employer straight off college.

If we're doing that, then there's a lot less of that tension and headbutting that can happen in organizations, especially when we don't see each other all the time in person. Some of the other elements along the be kindness is say thanks and say sorry. And negative feedback is one on one. Yeah. By the way, for those who aren't looking at the values you can go to about dot get lab.com values and you can see them.

But what I will share with you is that we have a thanks channel where people just post a thank you note whenever they are truly thankful for something. And that channel is constantly getting those messages in and it's reinforcing that engagement that you really want to have where people are working together collaboratively and assuming positive intent and so forth. But I will tell you, the negative is one on one that you mentioned, Lenny. That one's actually really important to me because you don't want someone to take a public message the wrong way. And so you take that conscious step to say like, hey, I got to give that feedback.

Let me do that one on one. And that means if it's in a public channel, it's even easier to assume the positive intent because negative feedback should be one on one. Yeah, I think the more you talk about this, we're going to talk about remote work. It all works together. Remote work almost forces you to be very specific in how you communicate, be transparent about all the things going on at the company.

Lenny Rachitsky
So this all makes a lot of sense. It's kind of this flywheel, I imagine that kicks in. I agree with you. I think you can be as transparent as we are without having our values. You can't have our values unless you're trying to put it sometimes into the remote work environment and vice versa.

David DeSanto
And I think they're all fit really well together. And its been the company continuing to grow and adapt them as weve gone from, again, one person to over 2000. You mentioned the value of short toes, which I was definitely going to ask about. Talk about what that means. Yeah.

So the short toes is about the work. Its not about you. And so if you have made something and youre getting feedback on it, try to realize that as someone trying to help you make it better and not a personal view of how you are as a person or as an employee. And so a great example would be I make videos occasionally and post them on slack and people can consume them and those are not always well received. I dont take that personally.

I take it as an opportunity to say what could have made this video better. So the next time I do it, it provides a lot more value. And I think that is really where that short toast comes in. Its about comment on the work, not the person. It's not, look what you did.

It's like this could have been better. This way. And if you're assuming positive intent, having short toes, you end up having less head butting. I think the implication here is that you don't want to feel like people are stepping on your toes, so your toes end up being short. Right.

Lenny Rachitsky
That's, like, where this comes from. I imagine it does. And part of that ties back to those other values. Right. You aren't your work.

David DeSanto
Right. So if someone's reaching over and doing something in your space, that's where if you have long toes, as they're called, you feel like people are stepping on you. Whereas if you had short toes, it's about contributing and so forth. It's so funny. I love the way of describing it.

Lenny Rachitsky
And it's the exact opposite of Uber, who's one of their values, I think was encouraging. Toe stepping, step on toes. Don't worry about it. Long toes. Our approach was to take the other side of the move fast and break things.

David DeSanto
Right. It's more about building that community, that trust, and everyone hitting the results together. Sounds like a delightful place to work. This is the happiest I've been in my career. Has nothing to do with my success at GitLab, though I'm sure that's part of it.

But it's also, I know that we're all working together for the common good. We're all trying to aspire those values, and it makes it a great place to work. And it's been the happiest I've been, and I've been here four and a half years. Wow. I love hearing this.

Lenny Rachitsky
And it's done great also, which is, you know, it shows that you can run, you can build a very successful business, and people can be very happy. What a rarity. Yeah. No, agreed. Are there other values that you find to be really core to the way GitLab operates that I haven't mentioned?

David DeSanto
Yeah, I think for me, it's the results and the efficiency ones that always hop out for me. And results is focused on results for customers, and that's why we do everything we do. Almost any company, that's really the case. And so how do we help our customer get their software ship faster, make their software more secure, give them better visibility into its usage. And having that focus on the customer is the number one thing.

It allows us to be a lot more empathetic with not only our customers, but the broader community. And efficiency is really about how do you get work done? And we try to drive responsibility down to the lowest level in the organization. I, as the leader of product, will set a three year, seven year strategy. But how we get there we leave up to the individual teams and that empowers them to be able to work efficiently and hit that high level of velocity that we talked about at the beginning of our chat.

I think for me those things come together. It's okay to ship something that you think is good, but maybe it's not great yet. Get it out there, get that flywheel working so you get the feedback from the customer and then iterate on it. And I think those two things together on top of what we talked about, the transparency and the collaboration, really help us be who we are today. Did you say a three year and seven year vision?

Yeah, we have on the website up to I think a 30 year mission. So it's, yeah, we have our mission, we have our vision, we have our strategy and we have our direction. Those are the levels at which we plan. What's the thinking there? Uh, the thing is to make sure we're always going towards where we think our customers are going to need.

And so to give you uh, an example, the three year strategy says our goal is become the first all Ops platform. And that essentially to say that we want to be the single source of truth for R and D organizations. And that's not just the people who are in R and D, like product Ux, engineering infrastructure, but for the teams that work around them, including legal, sales, product marketing, compliance and so forth. And so for us at all, Ops means it's everything that's needed to do that. And that's what we've built.

Not just source code management, code review and so forth or CI CD. We've added security and compliance, we've added enterprise agile planning. We're even starting to add service desk service management functionality in GitLab as well. For us, that's our long term goal. And then we empower the teams to get there the way they think we need to get there for their area or a new area that needs to be created.

For those who are not as familiar with GitLab, we're structured by sections. Those are areas like dev, Sec and ops. Those have stages in them that map to the DevOps toolchain like create, plan, monitor, verify. Then those have groups in them and those groups are that lowest group I'm talking about where we push down the priority and those will be the things that have our categories. So there's a group that's called environments that manages our Kubernetes integration, our deployment dashboard and so forth.

And that of course then all feeds back up to that top level vision. And so it's allowed us to be structured in a way that helps us ensure that the people who need to work closely together can across those groups. And it allows us to tie stages together to make sure we're hitting the outcomes that we need. And so that kind of is how that all then fits together. There's a group direction, there's a stage and section strategy, and then there's companies longer term vision and then mission.

Lenny Rachitsky
And I love that all of this is there in the handbook. Like you can see all this and then you can also see your cadence for revisiting it. Planning, team meetings, executive staff discussions. Yeah, and by the way, we also put thats all the way down to the group level. So on our marketing site, if you go to about GitLab.com, again, direction under direction is the product direction that weve set for the next year, and then that links out to those stages and groups so you can really follow along.

David DeSanto
And when you get lower into that to the group level, theyre linking to their epics, their issues and their tasks and our issue trackers so you can easily find them. We have, I think 72,000 open issues right now in our backlog and be hard to dig through all of that. Its people creating new ones, customers commenting on it just keeps the backlog really large. But you can go through that marketing site that we have and get to the thing youre actually interested in reading about. There it is.

Lenny Rachitsky
Im looking at it. I love it when someone joins GitLab and things dont work out. They dont end up being a fit other than them, not just like having the skills they need. What do you find is the most common reason for them not being a fit at GitLab? So I think it really kind of dials into the remote experience.

David DeSanto
Remote is not for everyone. And we'll have people who come and they're really great. They don't even have to be an underperformer, but they're missing that in office every day feeling maybe some, in some cases they just want to get out more and be not at their home or at the same co working space all the time. And so for them, that's really the thing. That's the thing for them is that they just don't feel connected the way they want to be connected.

And I'll be honest, I started at GitLab a couple months before the pandemic. And so for me, I got to attend to our sales kickoff. But that was only a third of the company. And it's not until actually in March this year that we're doing our first company wide get together where I'm going to get to meet some people I've never had the chance to meet in the four and a half years I've been here. And so I myself found that a little jarring.

Initially, I worked in a company that had a hybrid model, and I knew what that was like to have people not with you, but the fact that human connection can be missed more so for some people than others. And I think that's really what it comes down to. Yeah, you're right. They could not understand the technology. Maybe they're not fitting into the role.

But I think sometimes it's also, I just. I'm not enjoying the all remote. I'll call it lifestyle, because it is an adjustment to how your day goes and how you engage and all that. Sort of thing that makes sense. And that's a great segue to where I wanted to go, which is talking about remote work.

Lenny Rachitsky
So I think you might be the largest all remote company. Do you think that's true? We were pre the pandemic when we were around 1100 people. I think everyone now having recalls back to the office, we might be back to being the largest all remote. I've not seen a number yet from other companies.

David DeSanto
How much are they? Still remote. But I was during the pandemic. I think a google, an apple would win as the largest all remote. Right.

Whereas, yeah, now those people are coming back to the office, we might be back to being the largest. Okay. And I think even if you aren't currently largest, it feels like you will be probably long term because people kind of go there and then move back. Also, you, GitLab has been doing this way before school, way before everyone else started to try to go down this direction. Feels like there's a lot that people can learn from how to work remotely successfully.

Lenny Rachitsky
I'm curious, when you talk to other leaders and other companies that are trying to improve the way they work remote, what advice do you share? What tips do you have for making remote work more effective? Yeah, I think there's a handful of things that I really encourage people, which, by the way, I do get reach outs on. It's now called X LinkedIn, where someone will say, I read this in the product handbook, but I don't understand how, like, how this then applies in an all remote environment. And so I usually kind of say, like, if you're looking at being a remote for first culture, there's a couple of things you just need to understand and focus on.

David DeSanto
Things like be transparent. We've talked about that a bunch already. Focus on results, not hours. Sometimes people will get fixated on the, well, did you work the full 40 hours this week? Or, wow, you worked more than 40 hours this week.

What's going on? And instead it should be about, hey, we've agreed to an outcome and how we achieve that outcome. I think over communication is also key that I share with people. The best way to put is I have an executive coach. He's a great guy.

One thing he said was that if you think you're communicating 100% accurately, that's probably 60 or 70% for the other person and shoot for 150%. And so that's something that I think really helps in that environment. And then lastly, and we just talked about this is making time for in person events. I have my leadership team get together every quarter. Every other quarter of that, we bring in our director plus group and were actually doing that next month.

Last time we did it was in October. And see if you can find ways to get more part of the team together. I just shared were having our first company wide get together since the pandemic, but that doesnt mean that we had to wait that long. I got the product division, all 140 of us together across two different cities over the course of a month, to allow people to finally get to meet each other and make that human connection. I think when someone gets to do that, it makes Zoom feel a little less scary because you've actually seen and maybe hugged or shook the hand of that other person, and now you've got that same human connection that you may not otherwise get if you just never saw your coworkers in person.

Lenny Rachitsky
So the four pieces of advice you shared, transparency, spent a lot of time just being more transparent. Focus on outcomes versus quality, quantity of hours over communicate, and make time for in person within the outcomes bucket. A lot of people like, yes, it sounds great. We will focus on outcomes. We're not going to think about how many hours you're working.

It's hard to do a lot of times because in this sort of work, it's like, I don't know, this, like, what should I expect of this person to achieve? Maybe this bug takes a week, maybe took a day. Do you have any just advice to help figure out a nail? Like, here's how we can create outcomes that connect well with people actually working fully. Fixing a bug is kind of like more of a deliverable than like a business outcome.

David DeSanto
And so we try to make them things like, we want to have 60% of our customer base using this part of the portfolio. Now what do we do to get there? And that allows people then have a measurable outcome that does tie back to how many hours they're working to a degree. Right, because you're scoping it. But it's not about the ship.

20 features next month. That could take you more than a month, that could take you a day and a half if you break down your feature into tiny little features, right. Whereas it's more like celebrate the adoption, not the shipping. And I think that's really where we've tried to move. GitLab as a company has always been very focus on how fast we ship software, but that's not really how our customers use it.

It's more about what pain point or use case did we solve? And if you're now framing it in that way, you begin to move away from the deliverable and the hours and you get into the did the customer adopt it? What was their outcome like? What success did they have? What are you trying to do with that?

And by the way, I think that's the biggest challenge for people who move from a different role into a product division or product role is changing that dynamic to not be like the bits and the bytes, but be the use case, the pain point, because sometimes you actually dont need to ship a new feature per se. Sometimes you just have to make it more usable and then all of a sudden youre getting the outcome youre looking for. You said that a lot of people reach out to you and try to dig into some of the things you do. What are some of those most common questions people ask you and or where do people most often go wrong when theyre trying to work the way you all work? Probably the top topics that people reach out to me for the first is how did you get into product?

Because my background, I started as a software developer, had my own company, sold it for all those listening, just enough to pay off its bills. It was not like I independently wealthy now moved into security research and eventually into product. And it's like, how did you make that leap? How did you know it was the right thing for you? The next topic is the, I see GitLab has this very transparent handbook and this GitLab unfiltered channel.

What, how did this happen? Why are you doing it? And then the last one is usually, how do I know remote development is for me. And I make the jump and I talk about myself and my time in the office versus not in the office. And the things for me, I was very transparent with the team when I started 2019.

That I'm like, this makes me a little freaked out. I'm not sure this is going to be the thing for me, and I love it now. I can't imagine not doing it. I like to look back at that time when Gillard made the offer to have me come in and start leading a part of the product division, and my wife saying, you're going to love this. I don't know why you're questioning it.

So listen to your friends, your family, your partner. They can also help you with that sort of decision. This episode is brought to you by Paragon, the embedded integration platform for B two B SaaS product development teams are your users constantly requesting new integrations with other SaaS platforms that they use? Unfortunately, native product integrations take months of engineering to build, and the maintenance never ends. Paragon enables your engineering team to ship integrations seven times faster than building in house by removing the complexities around authentication, messy third party APIs, and debugging integration errors.

Lenny Rachitsky
Engineering teams at companies like Copy AI, Cinch, TLDV, and over 100 other SaaS companies are using Paragon so they can focus their efforts on core product features, not integrations. The result? They're shipping integrations on demand, which has led to higher product usage, better retention, and more customer upsells. Visit useparagon.com Lenny to see how Paragon can help you go to market faster with integrated today. That's use paragon.com lenny.

Let's do a quick tangent on that first bullet you mentioned of getting into product. What advice do you share with people? And they're just like, should I get into product? How do I get into product? I'm very transparent.

David DeSanto
That product is not necessarily the rockstar role that people think it is. You know, the way I tell people who are new in their career, maybe they're now an associate, intermediate, p advocate, lab, or other places I've worked that, like, if everyone's really annoyed at you, you're probably actually doing your job well, right? Product is the, like, the hub in the middle of the wheel, and the spokes are engineering, marketing, sales, legal, and so forth. And like, they can't do your job unless you're properly leading. And if you're pushing the boundary and they're all kind of getting a little like, are you really sure that's what you want to do?

Or, wow, that seems like a lot of work. Or, huh, I hadn't thought about it like that. You're probably doing your job well. And what I encourage them to do is think outside the box, not focus on the, I call it order taker or deli counter worker in the US where everyone gets a number and you just fill what that is. Think about what do all those requests together really mean and then do that and not take the orders.

And so thats kind of the advice I give is like its not the big sexy role that everyone talks about. Its actually a lot of work, a lot of discussion, sometimes tension and lead with the customer first and the use case and the pain point and then everything else will kind of fall in behind that. I completely agree. Its the most thankless, painful job there is. But also the best job.

Yeah, I mean I now have been doing it for a while. I can't imagine not being in product, you know. And so I'm glad I made the change and I've not looked back specifically. With PM's in a remote culture. I never worked in the remote world as a PM COVID happened after I left the field and started doing this crazy, weird job.

Lenny Rachitsky
Is there any tips you could share with PM's that are just struggling in a remote world where it used to be you just walk up to an engineer, hey, how's it going? How's this thing looking? Let's check out this design. Give me advice for just how to do those sorts of micro interactions and stay on top of things. Yeah, there's probably two or three points I would make.

David DeSanto
The first is if you're in an all remote world, your requirements have to be clear. Sometimes you can get away with, hey, I have a daily stand up or a couple weeks stand up. And then you use that as the opportunity to clarify something. You have to get it written right the first time or at least refined before you expect someone to work on it. And so as part of the interview process at GitLab, we do what's called a deep dive interview where we have the person try that out and they engage with another person or product who's playing the engineering side of that conversation.

And it really helps both us see whether or not the person has the potential to be able to work in that type of environment. Oh, this is part of the interview process. In the interview process, correct. And then it also gives them the opportunity to decide can they actually do it? And we've had candidates who go like, this wasn't for me.

And so that's the first thing is writing real concise requirements is really key. Just to understand before we move on. So in the interview, so what is it that they do? You ask them here, write the requirements for this feature that we have and then they kind of do a role play with as part of the interview with an engineer asking them questions about it. So it's actually they can write requirements for anything.

We're not using it as a way to get free requirements written up as part of the interview process. So like one candidate, their topic was they chose bicycles. And so it's like you're going to ship a new bicycle. Then like what's the epic, like the higher level vision and like what's the first milestone for us? Milestones or releases, but like that first iteration, what is the NBC for that or minimum viable change towards it and that then they create that after having like an hour zoom call with the person who's going to do it with them talking through it.

Some they then go write those things and they engage with that person. It's always someone in Prague. We don't pull our engineering team in to do it, but you know, they'll ask a follow up question on the issue and then let them rewrite something in the issue or they'll ask them a question maybe that causes them to define something better. And I personally when I was interviewing thought like this is a weird step, I'm coming in as a director at the company. And then I finished it and was like this is the best experience I think I've had.

Like this has to standard interview process. And so that all ties together as that 1st, 1st thing. The second thing I tell everyone is don't wait. Like if you're concerned or wondering something is going off the rails or you just want to know some things a on track, like don't wait until your next check in point. Whether that's a lot of teams here do a once a week stand up.

Like just ask a question on that issue or on the merge request or send a slack message right away. The last thing you want to do is have a 24 hours, 48, 72 hours cycle go by and a developer, your engineering manager is stuck or they've asked you a question and now you're leaving them kind of hanging and out not getting the answer they need. And that's where like to your point, you used to block up to a cubicle taps on the show and be like hey, what's going on? And now it's got to do that digitally and that is reach out. On the communication platform we use Slack.

It's like reach out on Slack or comment on the GitLab issue or merge. Request for this to work, the values again come up where you assume good intent. You be kind people because I think oftentimes the companies that PM checking in often feels like goddamn leave me alone. I just want to work on this thing. You just don't trust that I know what I'm doing.

Lenny Rachitsky
But I think with these values that helps avoid that. I imagine it does. And I think it also kind of feeds back into those remote work items as well. That it's actually good for you to over communicate, it's actually good for you to focus on. Is the outcome happening?

David DeSanto
Because you may end up in a situation where the outcome felt like it was really big, but three days into your one month long iteration which we have here, all of a sudden you realize this can be knocked down a week. Now don't just wait, reach back out and say hey, this is done. What's the next thing? Awesome. You mentioned Slack is a tool you use.

Lenny Rachitsky
What other tools do you find really interesting that people may not be thinking about for working successfully in a remote environment? I'm going to be biased here and say GitLab as a product we call it dog fooding, but we use it for everything we do. And thats become key because it makes sure that the product reshipping is going to work for our customer and our end user. But our policy here is hey if you have the idea, open an issue and then tag your team and then that issue becomes that single source of truth, then we can then turn that into an epic. Or if its going to become work for the engineering team, it can be converted into one of those types of issues.

David DeSanto
We do it that way. We also encourage people to use Zoom if you have too many back and forth. So like say like, like Lenny, you ask me a question, I answer it and you go but I understand, like you ask it a different way and then I answer it again. You're so confused. Like practically reach out and get on Zoom because maybe it's just something in writings not translating.

And so between those three main communication tools, that's what we use. GitLab is very anti email and a lot of ways obviously we use it for external communication, but internally it's on an issue, it's in slack or it goes into the handbook. If we've made a decision together, whatever you and I were going back and forth on, we think this is really important. We go update that part of the handbook or add a new handbook page or whatever needs to happen so everyone can benefit from it. So when someone updates an issue, do they get pinged in slack when people go check it out?

Lenny Rachitsky
It's not like an email. Yeah, it does. If you have your notification set up, any comment on something that you created will give you a new notification. The notification shows up in your to do list and GitLab, you can have it send you an email. I have it actually email me and it goes into a folder actually created.

David DeSanto
We use Google Workspace and if it's a, you've been mentioned to do it goes into one leg or gets one GitLab Gmail label. If it's a, hey, this is an issue that you're following that there's been activity on that goes into a different label, and that way I can quickly find the ones that I should be responding to. Got it. You talked about this handbook workflow. So I was reading a bit about that.

Lenny Rachitsky
So basically the handbook is like essentially treat us code. People submit pr requests to change the handbook, which essentially is changing the way the company's operating. Can you just talk about that workflow? There's a piece of update the handbook before you make change. Yeah, we talked a little bit about single sources of truth earlier, and one of those is the handbook for us, it's how the company operates.

David DeSanto
And so if you're finding something that's inefficient in a workflow or you're finding something is unclear, you're encouraged, even as part of your onboarding, to open up a merge request and make it right. And we sometimes make jokes in the product division, because when product managers are onboarding, the handbook feels like the product, and in their three week of onboarding, they may end up opening up bunch of merge requests for things like, I found this typo, I fixed it, or the sentence could have been worded better, so I reworded it. Now it makes more sense, right. But then theres also the bigger thing. So our product development lifecycle and that workflow and the framework that supports it is all in the handbook and its in the public handbook.

And so what that allows people to do is understand exactly how were going to operate for a milestone, that iteration again, where, hey, whats expected three weeks before, two weeks before from you. And then that allows some of those values we talked about earlier to come into effect, because now youve also have your em on the other side who also has something similar and knows when to expect something from you, when not to expect it from you. But then if there are bigger changes, we of course have processes for those that are defined great examples. If youre going to introduce a new category, which new part of the product, it requires mysign off to do that. If you're going to change that product development framework that helps you prioritize, that's going to be approved by myself, the CTO, as well as a lot of our directs that are involved in making sure we're delivering efficiently.

And so it can all be from a team perspective to a global wide that you can contribute to. What we have encouraged people to do is create their own mini versions of these things, because, like, I want to provide guidance. I don't want to give you direction. And what I mean by that is like, here's how I'd like you to prioritize. Here are the important things for the company.

But whether you make that one milestone is 100% new features, next one's 100% bug fixes, as long as you're achieving the outcome that we've said, how you get there is your own decision. And then those individual teams create their own mini version and references back to the bigger handbook. Got it. One more tactical question about remote work. What's your time zones policy?

Lenny Rachitsky
How do you align time zones and get people talking and not bothering each other when they're asleep? That's a good question. So we focus on asynchronous communication first. And so that means that, like, key decisions that involve people who are in a time zone where the meeting's not occurring, that waits until they're back online because they're the directly responsible individual or, or DRI, as we say here. We also focus on making sure the meetings that do happen are optional and are both recorded as we talked about the beginning, but also have really good notes.

David DeSanto
And so that way someone who's in a different time zone can, can read that and consume that when they're online. Our goal is to make sure we're as inclusive as possible. And so that includes even leadership roles. Our leader of our product portfolio for our SaaS platforms, which are GitLab.com comma and GitLab dedicated. He lives in Germany, but hes a leader at GitLab.

Lenny Rachitsky
Right. And his product team is in the US. I think hes got someone in South America, people in western Europe and so forth. And so that means that he may not be online when parts of his team are online, but if hes communicating asynchronously and giving clear outcomes and empowering the people who make the decision, its okay that they may not have overlap. I had a leader in our team who was in Tel Aviv at that point.

David DeSanto
I still lived in the middle of the United States. We had a half hour overlap in our day, but she was always highly efficient as a leader, and her team was also geographically dispersed, and they were highly effective. And to kind of give you the way ive looked at it, it means that you can empower the people around you so that you don't have to worry about those meetings that happen at weird times of the day for you. A good example, if I can't make it to a leader or meeting that is more APAC time friendly, because maybe that's where a lot of the team members are. I have leaders on the west coast of the United States who will take that call, and they're empowered to do that.

And so that's how we kind of deal with it. We are in 60 countries or over 60 countries at this point. We've had to be very careful to be inclusive, and that's what drives those things. Well, note taking, recorded meetings, that asynchronous focus on communication. I was just thinking, with all the recordings you've done, you could train an LLM to basically run your business at this point.

Lenny Rachitsky
I imagine engineers are thinking about that. I don't think anyone's thinking of that yet, though maybe they are. They have actually focused on how do you make a version of David, who's in every meeting and they're playing with some of the AI generated stuff, training it on videos that I did. We have a lot of fun at GitLab, but we also accomplish a lot, too. Is the takeaway from that.

Before I move on to a different topic, in terms of transparency or remote work, is there anything else that you think would be worth sharing or giving people advice on? The best summary of the conversation is make sure you're focused on communication, being clear, setting the clear steps like a framework for how to operate, and being uncomfortable with the transparency to make sure people can follow along with if youre doing those things right, remote is actually really great. I cant tell you how I feel today in a measurement thats not just like, im really happy about it, but like, I just felt like other places ive worked where we had to hire to an office, it really limited the candidate pool you could have. And now being all remote, we can hire anywhere for the most part and find the best person for the role. And that's allowed people to have the life they want.

David DeSanto
I don't live in the Bay Area anymore, and I think it's phenomenal. I can your family and friends that I grew up with then I can still work that tech job. And that's not just true for me, that's true for the gentleman I mentioned who's in Germany. We had an employee who is in Israel on the leadership team. We've had PM's, UX designers, technical writers, researchers all over.

And that's not just true for a product, it's true for everybody. Division at GitLab. Okay, so I asked a previous guest who worked with Yuheela Chu, who was head of growth, I think, at GitLab for a couple of years, and I asked her what to ask you and asked generally about GitLab. And she said that GitLab is really good at this breadth over depth strategy of building product. And that's been the key to success so far.

Lenny Rachitsky
Can you just talk about that and why that's been important? Yeah, so I did. I love Hila, by the way. I worked with her for about, it was a little over two years, I think the time she was here still talk to her occasionally over like LinkedIn messenger. Someone asked me if LinkedIn was dead, by the way.

David DeSanto
I think it's how I stay in touch with former colleagues and plan things out. I think they're on the way upswing. They're killing it right now. I think they are. I think that's a whole nother podcast on why that is.

But so Hill was right. So when she started in 2019, a couple months after me, we were very much focused on a breadth over depth. And that was so we could build out what is the Devsecops platform. The only way that we could truly help people from a platform strategy was to be touching the different parts of the Devsecops lifecycle or the SDLC, depending how you look at it. Now.

Last year we made the conscious decision to start to pivot to depth over breadth. And thats because we have a very broad platform. Today we are touching everything from business continuity planning and okrs, to enterprise agile planning and team planning, to coding, deploying, securing, monitoring the entire SDLC. And we now know theres key areas where we need to be really deep to help companies accelerate delivering software. And so in those key areas, they are source code management and things around that like code review, ID experience, remote development, CI CD, security and governance, planning and AI.

We know if those are really deep, the things that aren't as deep around them, we'll get that like a rising tide lifts all boats effect. And so we've pivoted in those queries now be depth over breadth. That's awesome. That's so interesting. Basically to create a wedge and gain traction sounds like initially those just do a lot of things but not be the best.

Lenny Rachitsky
And now that you're doing great and have all these products, the strategy has shifted to going deeper. This is a question a lot of founders always face. Should we go wide? Should we go deep? I know you weren't there at the beginning, but just I guess do you have any thoughts or insights from when it makes sense to go wide versus when it makes sense to focus?

Because usually the advice is just do one thing really great and that's the only way to win. And it sounds like clearly this is an opposite approach here. Yeah. And for those who are as familiar with GitLab as I know you are, we are the leader in our space. We are the devsecops platform.

David DeSanto
Not just us calling that, but user reviews. Analysts like Gartner and Forrester are calling that out. And so what for us it was was that I think when youre going really wide, youre trying to find your niche and what youre going to be really good at and what you can differentiate on. And for us what that was is where are the key areas that we can truly help someone accelerate delivering software? And over the course of those years, youre right.

We were still in a fast growth period when I started, and I was part of that breadth first when I started in 2019. But we found those key areas that we know if they are truly deep, they bridge to the other area thats deep. And maybe some of those areas dont have to be as deep as the things around them. And so good example is we started adding model ops functionality to GitLab in 2020. We focused heavily on adding depth to mlops, which is just one of the three pillars of that.

We know that data Ops will probably be needed at some point, but if we have a nice integration with data ops platforms, you can actually do all your ML ops work in GitLab and be successful at it with a lightweight connector or component that's there. That's what we've done is we've started really investing deep in those key areas, knowing that it's going to help those areas that we aren't investing necessarily as much and be still good enough and kind of touch on what you were touching on earlier about the getting to that good enough level. What's interesting is my last interview that I did was with the co founder of HubSpot Dharmesh, and they had the exact same strategy, breadth over depth. Also one of their core values is transparency. They share all of their financials with everyone, salespeople, everyone.

Lenny Rachitsky
They see everything. So it's so interesting that these companies are very different and have very different markets. But there's two really similar elements here, and I've worked out for both, I guess. Does that bring up anything? I don't know how much you know about HubSpot.

David DeSanto
I mean, I know the company and know some people who worked with me in the past that have gone there. I think what, what you're touching on is that there's some like product frameworks that are universal. You know, I occasionally will have founders reach out and ask the question, when do I make that pivot? And it's really like, have you found your fit in the market? And have you done the whole, and we'll use Jeffrey Mortars book, the crossing the Chasm.

Have you found that? And if you have now, you know, to put all of your wood behind that arrow to use a saying, right? And then you're able to get that differentiation and that market presence. And, you know, we go back to breadth over depth when we're looking at new investments. So it's not like it's left.

GitLab. It's currently part of our AI strategy that we're finding the spots for devsecops for AI makes a lot of sense. But we also know that if those five things I shared with you are leading, and we see them as leading in the industry, it's going to give us the ability to do more of that breadth over depth as we try to expand our SAM or try to go after a new market. Trey what's interesting with Dharmesh's approach, I'll share briefly, is their internal metric was if we're the top three product in any of the categories, we're doing too much. We're doing, we're trying, we're investing too much in these products.

Lenny Rachitsky
We don't want to be the top three of any of these yet. We just want to be fine because holistically it's going to be great. But just, I want to move on from this topic. But just, it's interesting. The approach you shared for GitLab of why going wide made sense was explore opportunities and then go big when you find something that works.

HubSpot strategy was our customers need all like the problem they have includes needs all of these many things. And so to solve their actual problem, it turns out we need to build a lot of things the way it is. And I imagine that was a part of GitLab's need too. Trey? Yeah, I'm sure.

David DeSanto
As we grew, I joined 2019 company started years before that. And I've just seen the maturity of the company as a whole and the growth and the need to focus on going deep in some really core areas. But like I said, we still go back to that same approach that they have. It just depends on what we're talking about and why we would make that decision. Cool.

Lenny Rachitsky
Okay, just a couple more questions. I know you guys are doing some cool stuff with AI. Let's talk about that. What's going on with AI and GitLab? I would say we've taken a very unique approach to AI as part of software development.

David DeSanto
And the reason why we did that is that GitLab has a very unique position. We are a true devsecops platform. A lot of our competitors that people talk about are like a developer platform or developer experience type tool. And we know that if 25% of the SDLC is actually creating code, 75% is not. And we want to help all those people around the developer be successful.

And to do that, we set ourselves like three core tenets or principles for AI. A couple of years ago, the first one was AI. Across the entire software development lifecycle. We want to help product managers, QA teams, ops teams, security teams also benefit from AI. If you make your developer 100 times more effective, it just means probably everything around them is going to break.

To make them more effective, you've got to help everyone. The next was we wanted to be both transparent, which fits into our conversation earlier, and be focused on privacy with AI. And that means that we tell you what models we're using. We're source available so you can see how we're using them. We tell you how they're trained.

And the privacy part is we don't use your intellectual property for training and fine tuning models. Our models are trained on data that is not yours, which means that you don't have more safety. And using that, I think other companies have, have shown why that's important with leaks of customer data and we want to avoid that. I think the thing that's interesting, as a source available open core company, we're actually trusted by more than 50% of the Fortune 100. So if more than 50% of those top companies trust GitLab to secure their intellectual property, we had to do that with AI as well.

Then the last one was AI efficiencies. We wanted to make it so there is a boost in efficiency. GitLab ultimate, our top tier has a seven x boost on productivity as from our customers doing a data analysis of their improvements. We want to take that to ten x as part of it. We think AI can give us that last little bit of a burst.

Now, I did mention a minute ago, and I'll stop there, see if you have questions on those. And then I just want to talk a little about where we're going with AI because I think it's pretty exciting. I guess the only question is just for other companies, other product leaders thinking about integrating AI, working with AI, any tips or lessons you can share that might be helpful on their journey? What I would say is we spend, and this is really the third tenant is finding the right model for the right use case. And good, I pointed out, and your question reminded me of it, what sometimes people will do is they go, oh, there's this really popular large language model, whether that's a commercial or open source, and they make everything fit into it.

And what you end up doing is actually reducing the quality that someone can experience with that feature that's leveraged by AI. I encourage people to find the right model for the use case. We have around 16 models we use today to make GitLab have its AI suite, which is called GitLab Duo. And we've done that to say like, hey, this one's really good at explaining and resolving vulnerabilities. We're going to use that for that use case.

This one's better at summarizing conversations in plan. Let's use that. And of course, code creation is also an area where if you're using the same model for inline code completion as you are for code generation, which is generating blocks of code, you're going to have a much different user experience. The things that generate blocks of code like functions, those actually take a while to run, they take seconds, sometimes 20, 30 seconds to fully respond. If you're just asking for the next couple lines to be completed, you're not going to wait that long.

And so you've got to have something that can handle that as well. And so we've done that, and I encourage everyone else to do that. If you go to the GitLab Docs website, you can select GitLab do and you'll see all the models we're using and you can start to understand. As you learn about those models, you start to see why we chose them. And it actually has allowed us to get closer and closer to that ten x that we're wanting to get our customers to.

Lenny Rachitsky
I'm pulling up that page but when you talk about models, are you talking about OpenAI's APIs for one and then Gemini for another, or is it internal models you're building? It's both. GitLab created open source and commercial. Today we've partnered with both Google and Anthropoc, and those are where the commercial models come from. Got to get it.

David DeSanto
Cool. Yeah. And then we actually acquired a company at the beginning of 2021 or 2022 named Unreview. And we have those models, those are the ones that would be GitLab proprietary. And then we do use open source as well.

We're actually looking at how we can leverage open source more and contribute back there. We have a really awesome AI model validation team. There are a bunch of AI researchers that I've studied, things like ethical AI and AI at scale, and that's allowing us to select the right model. Cool. I imagine part of this is because GitHub, your competitor, is owned by Microsoft and OpenAI, and you try to avoid that whole area, as I'm guessing is a part of this thought process.

Well, it's also about meeting those tenants. Right. And so our partners at their commercial, they need to meet that same requirement, that same vision, and both Google, as in Google Cloud and anthropic, were able to meet the privacy requirements we had. And those are really important to us. And so not every company that we could partner with could do that.

And so we had to be very careful as to who we partnered with. And so there's nothing about the companies we're not partnered with. Maybe we just haven't gotten to that point yet. Right. But that's part of being a partners.

You have to meet those same values for lack of other, to put it that we have base of our customer base, especially being trusted by more than 50% of the Fortune 100 makes sense. Okay. Going a totally different direction. Hela also said that you're a leader with a great sense of humor, and you often use your humor to navigate very high stakes conversations and executive negotiations. And I think a lot of people would love to have the skills.

Lenny Rachitsky
Do you have any advice or lessons other than just being David and being who you are? I guess. Any lessons for leveraging the skill, building the skill, how you apply it? I'll take the compliment from Helo. What I learned early in my career is that tension, especially very high, very tension driven conversations, can really kill productivity, kill morale.

David DeSanto
And then I found that if I can deflate that a little bit and kind of re add some, I dont say humanity to the conversation, but really disarm the topic that people are able to move forward. And I think thats what shes talking about. I will say that it probably is partially David being David to a degree. Maybe im leveraging the class clown experience of my elementary and high school experience and harnessing it for good now. But I think people can also do that and leaders who have seen me do that, like Hela, have learned that there's ways to do that.

If you can read the, read the room and the nuance and know whether or not it's okay to try to add a little bit levity, but I consider a value part of my tool belt now. Like a tool in my tool belt. And I'm glad she appreciates it. You know, not every joke lands, which I'm sure you're surprised, you're not surprised by that, right? So.

But yeah, I think it's, I think it's been good to help keep people moving forward. I love it. David, is there anything else you wanted to share or touch on or leave listeners with before we get to our very exciting lightning round? The big thing, I just like to let people know, if you're not familiar with GitLab or you haven't heard of GitLab in a couple of years, please go check out our website. We do everything across the software development lifecycle.

When I started in 2019, I was hired to add compliance and security to DevOps, which has made us a devsecops platform. Now we've gone from not just source code management and code review or CI CD, but true security and governance. That includes things like our remote development offering where now you can even get developers up and running in minutes on a software project, our enterprise agile planning capabilities and of course GitLab duo. We've heard from customers they've seen efficiency boosts of 50% and above by leveraging GitLab duo. I would say if youve not heard from us in a while or arent as familiar, please just check that out because we are truly unique in the space and its something I take a lot of pride in.

People said no ones going to want a platform, they want a bunch of point solutions and customers end up with 75 tools to try to deliver software and they adopt GitLab and all of a sudden they realize how much more effective they can be in collaboration. And even that transparency stuff. We talked at the beginning, I'm at. The website and even then I'm like, I still don't. I think it could do a better job telling me everything that you do so just to quickly summarize, so people know what are the bullet points of all the products and solutions you guys offer real quick.

We do everything across the software development lifecycle. Some areas are really strong. Those key areas are SEM code review, the what would be called the create stage in the tool chain CICD. We're industry leading there as well, security and compliance. So we can do application security scanning.

We can do things like software build materials and traceability and all the things that go around compliance and governance. And then of course we also have monitoring, whether that is value stream management and analytics. We can now do product analytics so you can have GitLab embed telemetry and then see how that app is being used. And we are currently working on expanding into observability and service management. But it is truly an amazing experience if you start with just an idea in our planning functionality.

Again, been awarded a leading enterprise agile planning solution and then take that from idea all the way to running in production and then learn from that and bring it back into the lifecycle. But ill take your feedback. Lenny, back to marketing and say, I was just on an amazing podcast, met a great guy, hopefully get to talk again in the future. And he said, I can't come back until the website's better, so can you please just make it a little bit clearer? Yeah, make the website clearer, software faster.

Lenny Rachitsky
That doesn't tell me what I need to know. I like this list you gave me. Okay, amazing. We're solving all the GitLab problems right here. There we go.

With that, we've reached our very exciting lightning ground. Are you ready? I am ready. What are two or three books that youve recommended most to other people? The two that I probably recommend the most are crossing the chasm by Jeffrey Moore.

David DeSanto
The current edition still has some things that are a little bit older now as examples, but the core of it is still very, very amazing. The other one is essentialism, and that is really about saying no to the right things and finding the thing that you really need to do to get the job done. And then not a book. But Jeffrey Moore created this. There's the critical core context framework, and if you've read both of those books, it really helps you frame in what is the critical part for your business?

What is the core, the expectation, and what are the things that you can say no to with the little bout of risk. So those things together I think are really powerful. Does he talk about that in zone to win or is that just something independent of his books? I think it was a class he created and it turned into chapter two in one of his books. I don't remember which book it's in, though.

Lenny Rachitsky
Super cool. That'll work. Next question. Favorite recent movie or tv show you really enjoyed? The most recent tv show.

David DeSanto
I really liked the Devil's hour. It's a BBC show. If you've not seen it and you like Sci-Fi crossed with like, I don't know, call it like a mystery criminal type show, it's really, really good. Movie wise, it's an older movie. My wife and I just watched the glass onion, really liked it and then there might be a little bit of a swifty in me and we watched the Arrows tour when it came available to watch and also was phenomenal.

I can't believe three and a half hours went by as quickly as it did. I also watched it and enjoyed it. Do you have a favorite interview question you'd like to ask candidates that you're hiring? Yeah. So I won't say there's a specific question.

It's so much a method. So I really like the star method for asking questions. If people aren't familiar with it, Google it. It's really powerful. And then how I then apply it to be my favorite question, depending if it's PM or it's someone in engineering or another department is giving them a scenario where there's a little bit of conflict or tension and having them work through that as an example.

I think those really kind of give you a sense as to how people can problem solve. Do you have a favorite product you've recently discovered that you really love? Yeah. So this is going to be a sad moment for me. My favorite new product is actually going away.

Artifact news. I love artifacts. Yeah. I think it's actually hit the point where it's hit its end of life, but it was revolutionary for news. I thought it was fantastic.

The things I use that are not the app that just went away that a little bit about. I've really gotten a love for superhuman, for email. It's really allowed me to make we're a Google workspace customer, more powerful. And most recently, I just started using the arc browser from the browser company. And I think it's fantastic.

I like how it organizes everything and archives tabs that they've been inactive for too long. And all these things that I think just took me from a single chrome window to five chrome windows, each with 100 tabs, and has made me more effective. Awesome. Also a huge fan of Ark the Josh Miller was on the podcast talking about how they operate. And superhuman is a sponsor of the podcast.

Lenny Rachitsky
So I love all your choices. Do you have a favorite life motto that you like to come back to or share with people that you find useful and worker in life? I actually have, like, two or three. I'll say them quickly for everyone. But the first is, like, as a leader, my first one is, your team's accomplishments are theirs, but their failures and misses are yours.

David DeSanto
And I, early on in my career, would have loved to feel like sometimes, like, my work was indigenous. Then praises from my manager, and as now a leader of a very large team at GitLab, it's been important. The one that my team laughs at that I say a lot, is, as a general motto, like, I'll go, like, just make it work. And people who are listening to your podcast might realize that's a Tim Gunn saying from Project Runway. And it was kind of like my way of taking on the work, the problem statement, and making it a little bit more like, okay, this is the thing you have to figure out.

Make it work. What is that? What does that mean for you? And so also lets me throw in pop culture, too. And the last one, and I'll say, I came up with this when I was back as an engineering director, was, you know, it's just software, so anything's possible.

You know, a lot of times someone will say, well, I can't really do that. It's like, well, but it is just software, right? Like, it's just. It hasn't been done yet, but it's. It's doable.

Lenny Rachitsky
David, this is so fascinating. The culture at GitLab is fascinating. You're fascinating. Thank you for making time and sharing all these stories and insights. Two final questions.

Where can folks find you online if they want to reach out and learn more? And how can listeners be useful to you? If you want to follow along with GitLab? We're at GitLab on basically every social. So go to any social media you can find us.

David DeSanto
For me personally, I'm Dee Desanto on LinkedIn. You just search for David Desanto. I should come up. I'm David Desanto on X, and as I mentioned earlier, I am David the beard on threads. I'm also trying to make blue sky a thing for me.

So if you want to catch me, I'm just d. Desanto on blue sky. And then how listeners can really help. There's really two things. Like, one, please share feedback with GitLab.

With my team. You can go and comment on our issues, our ethics, our tasks. It's all public, so flip things up. Give us feedback. That's always very helpful.

Just go to GitLab.com and you can see our code base. And then if you are someone who wants to contribute, like please open up a merge request, improve the handbook. Companies fork that handbook and then use it for theirselves as well. So you're not just helping GitLab, you're helping everyone as well as maybe contribute to the code. It can be as simple as fixing a small bug to adding a new feature.

Some of our favorite features are ones that others have contributed outside of GitLab. Awesome. Well, David, again, thank you so much for coming and sharing. Now I'm going to go watch this meeting you're in later on YouTube. Anyway, thanks for coming and thank you.

Yeah, thanks for having me, Lenny and love the conversation. Let me know whenever. I'll come back. Amazing. Bye everyone.

Lenny Rachitsky
Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review as that really helps other listeners. Find the podcast you can find on all past episodes or learn more about the show at Lenny's podcast.com. See you in the next episode.

The GitLab way: Kindness, transparency, and short toes | David DeSanto (CPO)

Primary Topic

Episode Summary

Main Takeaways

Episode Chapters

1: Opening Discussion

2: Deep Dive into GitLab's Transparency

3: Remote Work Strategies

4: The Importance of Kindness

5: Closing Thoughts

Actionable Advice

About This Episode

People

Companies

Books

Guest Name(s):

Content Warnings:

Transcript