Ep. #109, Less Context Switching with Esteban Vargas of Watermelon

Brian Douglas: Welcome to another installment of JAMstack Radio. On the line we've got actually Esteban from Watermelon Tools, so Esteban, you want to say hello?

Esteban Vargas: Yes, yes. Hello, Brian. Thank you very much for inviting me to this show. I am actually a fan of the podcast, so it's pretty fun to be here, yeah.

Brian: Yeah. I'm glad you reached out to share about what you're working on because, right after that, I said, "Hey, let's have you on the podcast so that way at least the audience can also share in this vision and what you're working on." Also, I think people might be interested in using the tool you're working on. But before we start, can we find out who Esteban Vargas is as well? What's your background? How'd you get here?

Esteban: I'm a software engineer based out of Latin America, in Colombia. I have been developing software for the past five years professionally, and very recently, to be precise, five months ago my best friend from college, Esteban Dalel, and I decided to start this company because we were working as contractors for remote, global companies with teammates in Latin America, but with teammates in the US, in Europe, in India.

Every time we'd switch clients, like every three, six or nine months, it was very hard for us to onboard ourselves to these new codebases every time. Yeah, even having like five, six years of experience developing software we always saw this thing as being something very difficult, and we started Watermelon, we started this company to solve that problem.

Brian: Okay, excellent. So what is Watermelon? I'm curious.

Esteban: Sure. So Watermelon allows developers to understand code contexts immediately. Right now we're doing that with an Open Source VS Core extension that allows us to hover over different lines and blocks of codes, and we automatically index information from GitHub and provide you with the most relevant information from GitHub for that block of code by running GitHub in the background so you basically save time of switching contexts between GitHub and the ID.

That's how we started, and then our vision is to add more integrations such as Jira, which is the one that we're working on right now, so that you can also get not only the most relevant PRs and the most relevant commit messages, but that you can also get the most relevant tickets.

Brian: Okay, excellent. So when you talk about information, I'm hovering over a line of code I see and I'm like, "I don't know where this came from. What's happening?" The tagline on your website is, "Go beyond, get blame." So I imagine if you're looking for the story, you're going to find yourself in the commit or the pull request and try to find context that way?

Esteban: Yeah, yeah. Basically we're saving the time of you looking at your code, do you blame yourself? Seeing who the responsible person is and then we do some common hashes searching for PRs and commits on GitHub, and seeing which ones are the ones that are relevant to the content that you're trying to gain. We basically save you the time of doing that, and with these version signs we're also going to save you the time of looking up Jira tickets and basically any other piece of technical information out there.

Brian: Okay, perfect. Yeah, I had a use case yesterday where I was writing some code on an Open Source project I have, and we take some contributions from some regular contributors but also some new contributors that might only do like one contribution. I was touching some code that was recently changed by another contributor and I was a little confused on how we got to this point where we had a function that was sort of working, but it was also over aggressive.

We basically just needed to write some tests around it to see how to contain this function and make sure that it's usable in other places. Pretty quickly I saw, actually in VS Code I have got the Git Lens and it just tells me who changed it. Shout out to George, George who's been contributing a lot to the Open Sauce project.

I was able to understand, "Okay, he's the one that changed it," but I didn't have the context of what PR that changed, what release that came in, how long ago this was. So I could see where this could be valuable, where if you're touching an old file, if you're touching a new file that's getting a lot of changes, at least you don't have to go back into GitHub.com, try to look through these random commit messages.

GitHub does have the nice Git Blame feature, but again that's going to be multiple tabs for you. Sounds like it's pretty useful, so I'm curious, how did you... You mentioned that you and your co founder were doing contract work for global companies, how did you get to the point where you wanted to solve this problem?

Esteban: Well, first of all, starting companies is very fun and starting a startup is very fun. We had actually done it in the past and we knew that we actually wanted to do it again once we had the right co founder and the right team. We just wanted to let it be.

We knew that the right problem would just come to us eventually, and we definitely found it doing remote contract work. It's very hard for us engineers to become productive immediately in a codebase and that was the main motivator to want to build this.

I also wanted to mention that you mentioned Git Lens, those guards are great. We actually started competing against Git lens. We're actually now seeing ourselves more as like a complement for Git Lens, and we're starting to show value on other parts of the interface of the VS Code UI, and secondly we're starting to show different information in different ways.

We're now in a new category of VS code extensions, but yeah, I guess that we were only in this space of individual engineering effectiveness.

Brian: Okay, cool. Then you spent some time working on this idea in an accelerator that was based out of Chile, which is Platanus Ventures. Do you want to talk about how that experience was, going through an accelerator?

Esteban: Yeah, absolutely. So the story was that back in late January this year, we were considering the idea. It was Miami Hack Week, we flew all the way from Colombia to Miami. It's actually a three hour flight, it's not that far away. We had already agreed that we wanted to build a developer tool because we were both developers and we thought that the best founder, startup, figured we were there because we understanding launches of developers pretty well because we have been doing this for years.

We said, "Let's go to Miami, and quick we'll have a bunch of developers physically close to us that can help us validate whatever idea we conceive there." We went to Miami Hack Week, we conceived of the idea and everything. Meanwhile, while being there, we applied to Platanus Adventures. They replied super rapidly, we had a bunch of codes with us and to our surprise a week later we came back to Colombia and we got an offer, $100K for 7%, they posted a money safe cap.

Yeah, they were super helpful. They were kind of doing the Y Combinator philosophy adapted to the cultural characteristics that we Latinos have. For example, the program runs in Spanish so it's for Spanish speaking founders. But they have a super techy vocation, these guys have started software agencies, cryptocurrency exchanges in Latin America, world, managers, et cetera.

So we were surrounded by a lot of software developers who are working at these companies that also became the early adopters of our product. Yeah, it was super helpful to get that initial instruction and that initial mass of users who stick to the product.

Brian: Yeah. I imagine the focus, I'm not sure Platanus Adventures has been around, but-

Esteban: Oh, two years ago.

Brian: Two years ago? Okay, perfect. Yeah, because I enjoy seeing the growth in Latin America for startups in the ecosystem. Obviously everybody looks at Rapi as the shining success story of startups, and I think famously it came from Startup Chile if I remember correctly. But yeah, it seems like the growth in the ecosystem is definitely needed because obviously not every Latin American country is at the same level as far as footing and economy and stuff like that.

I know Colombia has a very strong tech scene because I've been down there for JS Conf Colombia, is what it was called. So I was curious, are you involved in the local community in Bogota? Have you made it to Medellin to see the community there?

Esteban: Absolutely. Actually, I'm from Medellin. It's my home town, I'm moving back there in January after so many years living in Bogota because that's where I went to college with my co founder. Colombia is a country of a lot of engineers, a relatively growing English speaking population which allows us to work remotely for the US and Europe as well.

I think we just needed the Rapi phenomenon. Rapi is based out of Bogota, Colombia. We needed the Rapi phenomenon to make people start companies and it happened, and yeah, people are starting a bunch of startups now.

Brian: Yeah, it's amazing to see too, because we see this in Nigeria with the payment platform-

Esteban: Paystack?

Brian: Paystack, yeah. Paystack and Flipcard in India. These up and coming rapidly growthed startups. What usually happens is you see it with the PayPal mafia, is what most people point... the pinnacle point of where Elon Musk and Peter Thiel came out of. It expands growth into other areas as you see more of that VC funding looking for the next Rapi and I think there's a lot of excitement.

I personally love working with the Colombians. Hablo un pocito Espanol, so I enjoy going down there and practicing my high school Spanish as well. But yeah, I honestly think I'm bullish on Latin America and I think that folks should definitely give it a look for even hiring remote, as you mentioned. So what sort of companies did you work for prior to Watermelon? What was the problem you were working on before Watermelon?

Esteban: First of all, gracias. You should definitely come down and look at some talent. Fun story, my co founder's first job was at Rapi, so you could say that Watermelon is part of the so called Rapi Mafia. We're the newest startup in that mafia, but yeah. Prior to Watermelon, the last company that I contracted for was this company called Cerebrum. Awesome guys.

Affixing company in startup, distributed globally, again it had teammates in India, Brazil, in Germany, in the US. These guys were building blockchain verified credentials for COVID tests and vaccines so I learned a little bit about that topic over there. It was awesome. Then before that I was also contracting for various companies in the US in the logistics tech sector.

Brian: Okay, very cool. I wanted to take a step back then to Watermelon and try to figure out... You mentioned Jira. What is the future roadmap? What's next? Centralizing all technical knowledge on the platform, what does that look like in one to two years?

Esteban: Yeah, the big, hairy, ambitious goal is that you're just going to hover over a line of code and we're going to provide you the most relevant piece of information whether it comes from Jira or whether it comes from GitHub or Notion or Headache App.

It'll then do pager duty if you're a bigger company and observability, if it's an important thing to you. We're just going to provide you the most relevant piece of information so that you can understand everything about the past of that line or block of code.

And now to be very specific, we have this Open Source which is this GitHub with ES code integration, which is our free Open Source product. Now we're developing our closed source, which we're going to use to leverage our open core business model. In that closed source we're going to develop a series of dedicated endpoints, starting with the Jira ones that I'm mentioning so that you can basically get this relevant information from Jira without having to leave the IDE.

Saving this context switch for developers is what we believe will save companies years of productivity at scale once we do it with various companies because the context switch for developers is very expensive.

Brian: Okay. Yeah, very cool. It's interesting and I'm curious, VS Code. VS Code is something that I spent almost no time paying attention to up until the last couple years. It's surprising on how useful the platform is, and even for your product which I understand is an extension today. But what a great place to incubate your idea and your tool as an extension, knowing that... Off the top of your head, do you know how many users of VS Code there are? It's got to be millions.

Esteban: Yeah. I believe I saw somewhere there are 10 million monthly active users.

Brian: That's amazing. That market, the TAM slide on your pitch deck has got to be amazing.

Esteban: Exactly. I believe it's 21 million who have downloaded it, but 10 million are monthly active.

Brian: Wow, that's even more amazing.

Esteban: And regarding the TAM slide, yeah, there are 27 million developers developing software professionally in the world and we're going to be something like 50 something million by the end of the decade. So yeah, it's ever growing.

Brian: Yeah, absolutely fascinating. Yeah, cool. Well, last question, how can folks get started with this? Where do we go? How do we sign up?

Esteban: All you got to do is search Watermelon on the VS Code marketplace search bar that's inside the IDE.

Brian: I imagine that's a unique word, not a lot of Watermelon tools in the VS Code extensions?

Esteban: Absolutely, absolutely. There's a Watermelon color theme or something, but yeah, you'll recognize us because we have that smiley Watermelon logo that's very shiny. It's worked very great for us. Yeah, you can just do that and get started. It works both for private and public repos.

Brian: Cool. Honestly, I might go ahead and install that today. Maybe I can get unblocked, because at the moment I don't actually jump into code as often as I want to or I used to. So as things change around me and between my PRs, definitely very helpful to understand the context of the code I'm changing around me, so very cool. Folks, definitely check out Watermelon Tools.

I do want to transition us to picks, these are things that we're jamming on. Could be music, could be food related, technology related. If you don't mind I'll actually go first, I have a pick. It's a quick technology pick and it's called Dashi Base, Dashi Base is a layer that you can put on top of your Supabase database. Supabase has been on the podcast, a previous episode I don't have off the top of my head.

What I'm getting at is Supabase is an Open Source, Firebase alternative, it's built on top of PostgreSQL instead of a No SQL experience. With Dashi Base, what they've done is they've put a layer on top of your Supabase instance to then build an admin dashboard pretty quickly. I have one non technical person on my team who may or may not need to edit contexts and data. Currently what we're actually working with is we need to identify the most popular repos on the Open Sauce platform, and to do that we have a static table, basically, to add repo IDs to say this one's popular.

We want to automate that in the future, but for now, before we develop that algorithm to identify the most popular ones we've just been manually choosing them, based on what we've known. That experience can now be applied to Dashi Base where anybody on the team can interact with the database, specifically specific tables that we'd have to unlock to everybody else.

We don't have any personal information inside the platform, but if we did we can now limit it to, "Only you get access to this table based on this login." And it saves us some time because building internal tools for just your team to work with, it's either going to be a really underwhelming experience because you just do it, no design, just enough to make it work.

Or you can sign up for Dashi Base and it gives you a nice, beautiful... I'm not sure if it's a tail end, but a beautiful UI to interact with your tables. So definitely check it out, Dashi Base, probably have them on the podcast sometime in the future because I'm enjoying the experience so far. Esteban, do you have any picks for us?

Esteban: Does it have to be developer tool related?

Brian: No, no. Let us know about something local in Bogota so when people do visit, they can seek it out.

Esteban: I actually have a pick, I actually have a pick. If you are staying for a medium or longer term stay, I recommend something called Dos Comemos. It's basically prepared groceries that are healthy, delicious and relatively cheap. They get delivered to your home very fast, very conveniently. It's healthy and it saves you cooking time, so if you come to Colombia then look at Dos Comemos.

Brian: Say that again, Dos Comemos?

Esteban: Yeah, Dos Comemos. Yeah.

Brian: What's the English translation for that?

Esteban: Like, "We all eat."

Brian: We all eat? Okay, okay. I was getting there. Comemos, yes. Well, perfect. I'll definitely check out Dos Comemos when I'm down in Colombia. I personally love the country, a great experience. I only spent time in Medellin but would love to explore more of the country. To be quite honest, every time I go to a Spanish speaking country I get better at my Espanol so it gives me more confidence every time I walk into Colombia and other countries.

So I'll definitely hit you up next time I'm down there, and for this I appreciate you coming on and talking about Watermelon Tools. Folks, check it out.

Esteban: Awesome. Thank you very much, Brian.

Brian: And keep spreading the Jam.