Ep. #110, Online Whiteboards with Shin Kim of Eraser

Brian Douglas: Welcome to another installment of JAMstack Radio. On the line we've got Shin Kim. Shin, welcome.

Shin Kim: Thank you so much for having me. Very excited.

Brian: Yeah, before we get into what you're working on I'd actually want to know who is Shin and how did you get to this point?

Shin: Sure. My name is Shin, I'm based in the Bay Area and the founder of a company called Eraser. We make an online whiteboard for engineering teams. My background actually straddles finance and engineering. I studied computer science in college, and data science in grad school. But I started my career in finance, working with tech companies in the Bay Area.

I worked on some pretty cool projects like the IPO of Match Group, which is the parent company of Tinder, then also in VC investing. But a common thread throughout my career has been a passion for tools for thought, whether it's Excel or PowerPoint, in finance, or IDEs, or Jupiter Notebooks. I've always been fascinated by mastery of these tools to help us think through complex ideas as well as communicate them to other people, and that's what led me to start Eraser.

Brian: That's awesome. I actually have a finance degree, and then found my way into tech, which is-

Shin: Wow.

Brian: Yeah, I've shared this story publicly a few times. But yeah, I'm always fascinated by FinTech, I've already sworn off not going into that realm and touching that because it's a hard problem and it's for not the faint of heart, and I'm enjoying working the dev tools space.

Shin: For sure, for sure. For me, I think that's reinventing Excel is probably the idea that I've sworn off. It's this trap that people in finance that come up with tech kind of fall into.

Brian: That's funny, because how I learned how to... Well, I knew how to code from copy and pasting in high school and in college and stuff, but Excel is what got me into writing code full time because I was doing sales out of college with the finance degree, and the operations of setting up deals and doing macros inside of Excel. I learned all that to excel at my job, no pun intended, and really went down the rabbit hole of learning how to properly code and build web apps. Yeah, it's a meandering path and I'm super glad I made it here.

Shin: Yeah, I think it's a good gateway actually to programming and things like that.

Brian: Yeah, for sure. So you mentioned Eraser, that's what you're working on today, do you want to give us the full pitch? What's the goal with it?

Shin: Sure. So Eraser is a whiteboard for engineering teams. What the product is we combine a markdown note editor with a infinite canvas, and we introduce a lot of engineering friendly primitives like code blocks, cloud icons, diagram as code and keyboard based diagramming, which differentiates it from the general whiteboards that you might see out there.

We are particularly more relevant in this remote first, hybrid world, and the two main types of use cases are really the realtime, which is brainstorming, ideating, as well as system design interviews. And on the async side we have planning documents or documentation architecture, documentation documents that Eraser is used for. We launched in March of 2021, so a little less than a year and a half ago, and over one million people have used Eraser so far so I'm really excited by how it's going.

Brian: Wow, congrats. That's quite the feat. But yeah, it's also a very great problem to be solving too as well because as folks went remote, I think one of the jokes I made early on in the pandemic was how Google was doing interviews, because missing out on having the whole whiteboard.

Turns out there's a lot of whiteboarding tools out there as well, and I discovered them all because I do a lot of live coding on Twitch, specifically, and tried to basically unblock my design architecture and planning live on the air by also collaborating with other folks. It sounds like that is one problem that Eraser is solving, but I'm curious as to what's the other stuff that Eraser is solving as far as the feature suite and stuff like that.

Shin: Yeah, I think even in that live collaboration use case I think there's actually a lot of unsolved problems, I think, compared to when we were all collocated in the office and inside of a conference, together. We do a lot less drawing and I think it's because you have to open a whole app, you have to send a link, you have to ask, "Are you there? Can you see me?"

And then you start drawing and even then it's not like having a marker and having a go at a whiteboard. There's a learning curve, and I think it's exacerbated by the dominant whiteboard softwares out there being so general.

They're trying to be everything to everybody, and so on their apps, on their primary interfaces you'll see like 30 buttons. Whiteboards are supposed to be simple, but the reality of the apps today is that it's not.

I think the big reason why that's the case is because it is general, and that's why I'm trying to build something that's specific for technical users, focusing on concepts that are relevant and also introducing ways to do common things that yo have to do as an engineer or a technical person really quickly. We can go into some of these a little further, but we have slash commands where it's a universal launcher to put in, say, a Kubernetes icon or an S3 icon, and diagram as code which gives you a DSL to really quickly generate sequence diagrams and ERDs. Super painful to create if you're doing that from scratch.

Then even for flowcharts, just one click drawing of flowcharts or using keyboard shortcuts. So I think the first part is focusing on making things go faster, being really able to... You mentioned your Twitch livestreaming use case, if the tool slows you down it's kind of a conversation killer. If you can't really draw at the speed that you want to communicate, I think a lot of people just abandon the tool and they just use another medium.

The second thing is really process, I think as we switch onto remote work companies are needing to adopt different processes and async culture, a write in culture is often talked about and I think we fit into that. I think it's a continuum, if you think about ideation and brainstorming and then planning and receiving feedback, and then documentation once you're done with your task.

There isn't really, I think, a good process that every company has gravitated towards, saying, "This is the best way to do it." I think often tools do shape your processes, and each company has a different way of ideating and brainstorming and planning.

I think that was maybe less of a problem before the pandemic because the in person dynamic solved for a lot of the gaps in the processes, but I think they're becoming much more apparent in the post pandemic world where you do need specific processes and tools introduced in order to make that explicit, for teams to communicate better.

Brian: Awesome, yeah. I had alluded to this as well, during the pandemic I had gone through the gambit of quite a few different tools. It was more like a situation I was not really working on a specific team, but I do have the regular chatters on Twitch that show up consistently and every time I had to go whiteboard and map out, "Okay, this is what we're going to build, this is how I'm going to approach it, here are the features," it was like, "Oh man, what tool am I going to use?"

I know Google has Jam Boards and there's Miro that came out of nowhere, kind of a huge company. FigJam is the one I kind of centralized on mainly because I already have Figma design templates. Which, I'm not a designer, I just happened to learn Figma during the pandemic. I think the one thing you had mentioned that really was A Ha for me was having quick shortcuts for engineers to model things out.

It kind of kicks the keg for all those things that I've mentioned, because as much as I'd love to put a post it note in a FigJam, I don't want to do that anymore to architect what I'm trying to build, but post it notes are easy squares for me to grab.

Shin: Exactly. I mean, if you were to really boil down the first generation of successful web whiteboards like Miro, which is the 500 pound gorilla in our space, and I think FigJam is going head to head against use case... The killer use case I think is really sticky notes, and if you look at their interface it is often the most prominent button, it is often very much emphasized and so I totally empathize with you there, that if you're using one of those tools that's kind of the main tool that you're going to be using to communicate.

If you think about it, sticky notes are great but it's not always the thing that solves your problems for engineering communication, and so that is a paradigm shift if you come from those apps to Eraser. We really try to make it easy for you to do the every day things and do it in a one click or a two click way.

I mentioned the icons earlier, all you have to do is press the slash key and then type in S3 AWS, hit Enter and then you get a icon. And so it's simple things like that, but then again because these whiteboards have grown so sprawling in their use cases that they support, there's menu trees and carousels you have to go through to find the thing that you're looking for as an engineer. Again, that just doesn't work very well in a live conversation.

Brian: Yeah. The wheels are turning right now. I haven't been as public around this, but I'm currently writing a book for Manning and part of that process is you have to do some figures and diagram and essentially whiteboarding, but inside... What I've been using is Keynote as my tool of choice to make squares pretty quickly and shapes.

But now I'm thinking, "Wow, this would pair really well with the book that I'm writing, to diagram all these examples and then ship screenshots." So what's the collaboration look like with Eraser? So I have my team join, can I export some of these whiteboards and import these in my documentation?

Shin: Sure, yeah. Collaboration is a big piece of what we're trying to solve at Eraser. I think there were traditional diagramming tools, maybe like Draw.io or Lucid Chart, and less of a focus on collaboration or were built from day one as a collaborative tool. There's first of all the real time collaboration, all you have to do is copy paste the URL, send it to somebody and they can immediately join.

I think that's one of the differentiators as well that we actually don't require collaborators to sign up or create an account to be able to join and make edits. So it's super easy, whether it's in a interview or a live Zoom call, you can just send a URL. The other way is commenting as you would do on Figma or Google Docs. You can just leave async comments if you're soliciting feedback in a more async way.

Brian: Okay, excellent. I have a use case, literally tonight because I'm actually creating some diagrams. I've never liked using Keynote for creating that. But it's like once you have the hammer that you can solve the problem with, I've probably been using Keynote because of doing talks and conference talks and stuff like that. I can create a quick diagram inside of Keynote, so I guess I can do that same thing, grab a screenshot and share that inside of a markdown file on a GitHub issue or something.

Shin: For sure, for sure. Yeah, I think you're not in the minority, even, when you say you use Keynotes.

Just talking to a bunch of engineers, what their favorite diagramming tools are. Everybody has a tool but nobody has a tool that they love, and a lot of times, more often than not, you'd be surprised to hear Apple Keynote or Google Slides just because it's there and you know how to use it, and for Google Slides it's realtime collaborative.

But I think what we want to do with Eraser, because we're use case specific, I mention things like keyboard based flowcharting but we want you to be able to build that muscle memory so that you get even faster, right? I think with Keynote or Google Slides the diagram building is not really going to get that much faster. But with us, for example, we have a set of shortcuts that allow you to draw an entire flowchart without ever having to touch the mouse.

Draw things, change shapes, move around, and again that's just not something that you'll get either in a presentation tool or a generic whiteboard that's not catering towards technical users.

Brian: Yeah, for sure. Yeah, I definitely see the use case, and probably we do spread planning every other Monday so I think next time I spread a plan, I'll forego the Keynote and actually pull up Eraser to give it a go. I've talked about embedding this, taking a screenshot and embedding this into a GitHub issue. I'm curious, where do you go from now? How do you integrate this in a developers workflow? Do you have an integration with GitHub?

Shin: Sure. That's something that's very front and center in our mind right now, in terms of our product roadmap. So far we've been focused on the writing and drawing experience, the content creation experience because I think that has to be really pleasant and just polished. I think where we go from here is really embedding ourselves into the workflows of teams in a more deep way.

I think the first place to go is definitely source control, so with a GitHub or a GitLab integration what we're looking for is the ability to be able to create diagrams and mark down notes, and then save it into source control right. Our format, which is markdown as well as you mentioned diagram as code, it lends itself well to be committed to source control and you can run diffs and things like that because it's not a binary format.

So being able to save documents inside of source control, I think is number one. The other things that also become really interesting that you're able to do is if you create a link between a file or a code snippet and an Eraser doc, and you know the doc is documentation for hub architectures or how the code behaves.

When there is a PR that changes the behavior of a code or a file, you can run CI actions to validate the documentation and run alerts. I think going from where we are today and the ideation, brainstorming and planning world to even into the documentation world as we build these integrations. Obviously I think with GitHub there's also the comp onboard, the issues, and being able to bring in diagrams into those workflows too. I think it's going to be pretty interesting.

Brian: Yeah, I'm excited to, one, start using this thing. But also, two, seeing what that integration looks like. Always happy to nerd out on that, and seeing how closest to GitHub and building integrations in the ecosystem as well. Previously at Netlify, I'm happy to share notes there.

So I'm curious, is there anything else you wanted to cover as far as Eraser and folks? How can folks get started? What's the mental model of folks reaching for this the first time? What are they probably looking to accomplish?

Shin: Sure, they're probably I would say in a conversation and they're struggling to really communicate an abstract concept via just words. I think that's a really good time to grab Eraser. All you have to do is go to TryEraser.com, click on the Try Now button and then you'll immediately, without any signup or anything like that, you'll be sent to our app and you can share the link on a Zoom chat or a Google Meet chat, and you can start drawing.

I think that's usually the magical moment of that frictionless experience of not only getting the other person into the same space, but also the ease of use when you start using Eraser. There's no tutorial, no onboarding.

We try to make it fairly obvious and intuitive and I think once you have a couple of those under your belt, that really easy experience of being able to express yourself in a way that's visual and not just limited to words, I think folks slowly start moving onto more realtime use cases like interviews, as well as eventually planning docs and documentation too.

Brian: Yeah, indeed. Well, definitely going to try it out. I'm actually going through your templates. If I created a board without signing in, how long does it shareable? Do you guys delete that after the fact?

Shin: No, no. It's forever, it's persistent.

Brian: Okay, nice. It's like the GitHub, it's better than GitHub Gist at this point. Amazing.

Shin: Yeah. It just becomes part of history, so you might want to hold onto that link too though.

Brian: Yeah, might be harder to find. Yeah, I would say that the getting started, I'm a big fan of onboarding and shortening the steps to get to whatever the Ah Ha moment is, and the fact that I can just hit Try Now and I'm already starting to markdown on a board and I can start creating diagrams. Also the fact you give me some templates to start with, now my brain is like, "Oh cool."

The first thing I'm going to do right now is create an org chart for my team to help because we're currently hiring and we basically needed to fill in some gaps, and I need to clearly state what gaps we're trying to fill. But after this call, I'll create an org chart and maybe I'll even share it with you and you can recommend some engineers.

Shin: For sure, you'd be surprised how often teams use us to create organization charts and team structures. Those are popular concepts.

Brian: Awesome. Well, speaking of popular, I know this is going to be really popular after this episode. People are going to be signing up. Definitely try it out. Thanks, Shin, for coming in and talking about Eraser. I do want to transition us to picks, so this is what we're jamming on, things that keep us going. I like to call them, cleverly, Jam Picks and it could be music, it could be food, could be tech related. All of the above is pretty good and if you don't mind, how about you go first since you came in pretty prepared?

Shin: Sure. My pick is we're currently going through a marketing website overhaul, and so spending a lot of time in a tool called WebFlow, which is a WYSIWYG website builder that's powerful and I think very popular. But I think the beauty of it is that anybody can go in and make edits to the marketing copy, but it's not over simplified. So you can do a lot of powerful things.

In our upcoming website we're going to have 12 videos on our landing page and we're going to have a gallery site where we actually embed the actual product as an I frame onto the marketing page. And so it's powerful, at the same time because it's easy to create one of these things and there's lots of web lord developers in that ecosystem who you can hire for, I don't know, 5K, 10K to do an entire website, I think you experiment more because you know that six months down the road you can just do another website. I think that's just a new paradigm that WebFlow has enabled.

Brian: Excellent. Yeah, I've considered WebFlow multiple times and I think our next homepage is going to be WebFlow, to be honest. What I do like about it is that the designers can lead, so there's no barrier of, "Can we make this happen in code?" It's actually, "Can you make it happen in WebFlow?" Is the answer I want you to answer, so that's the one enticing thing.

All of these landing pages are amazing and I don't have the thought to answer a lot of those questions as an engineer, and I love the fact that you can unlock a lot of stuff. In scary ways, you can remove engineers from the equation to get something deployed, which I want to keep engineers happy, keep the dream alive, but also we do need to unblock stuff and ship.

Shin: Right, right. I think it allows engineers to focus on other things, right? Whether it's product or devops, rather than on the marketing website.

Brian: Excellent. Any other picks?

Shin: Second one might be another software, which is Linear. Another one that we use a lot internally, it's a next gen ticket management tool and we initially adopted it because as somebody who is building productivity tools and we like to think of ourselves as craftsman. We saw Linear as raising the bar in that regard, and it's been true to its reputation as we've been using it. Whether it's performance or keyboard based flows, which we ourselves are a big fan of at Eraser, the little details.

I'll give you an example, if you paste in a Loom link in an issue it'll automatically unfurl into an embedded video player without having to install a Loom integration or anything like that. It just knows to do that. And so there's been a lot of delightful, surprising moments and I think that's what we aspire to infuse into Eraser as well, so it's been a great opportunity to be inspired.

Brian: Yeah. We actually were talking about Linear internally. We have a small engineering team so not a lot of tooling we need today, so we're still mainly just using all native, GitHub features as far as product management goes. But I do like what Linear has done, and they seem to have really exploded their feature set in the last probably year, to be quite honest, because they've been around for a little bit.

But yeah, definitely something I need to check out and see, try to stay up to speed. And learning about Eraser today, Linear, WebFlow, these are all things that I think at the end of the day the developer tool space is always unknitting of ways to unlock and 10X folks. Because the illusion of the 10X developer, I think it is a myth for solo engineers, but you can 10X developers by removing a hurdle out of the way so whether it's your design system or whether it's productivity, that's how you 10X.

It's not actually getting people to work harder for you, work smarter. Cool. I do have two picks. One pick is actually what I just had for lunch. I've had this go to, whenever I feel like I need an extra push in the healthy realm my favorite salad is kale, pumpkin seeds and Parmesan with oil, salt and pepper. It's not too bad, a little lemon juice. It makes me feel better about maybe the chicken wings I had the night before. So when I feel like I'm going down a bad path I can course correct by eating kale at least for the rest of the week.

Shin: Is it homemade or are you ordering it from a restaurant?

Brian: I'm homemade. There is a restaurant below when I used to work at Netlify in the Dog Patch and it's the La Formagerie. And it's a cheese shop that sells wine as well, but they also do lunch, and this is a salad that they created which is kale, Parmesan, pumpkin seeds and then dried figs with some bacon. That's my go to. I don't pick salads for fun, but this one is pretty good.

It's pretty hearty and it's got enough in there that it makes me feel good about myself. So I highly recommend y'all, I don't really have a great recipe or if you are in the Dog Patch or if you're in San Francisco rather, La Formagerie. Which, I'm sorry, my French, the accent's not great. But order this salad, you'll enjoy it.

The other pick I have is I'm a shoe fanatic, I'm not crazy like currently in the shot I don't have shoes behind my head right now on camera. But I have been a frequent visitor of StockX and I like finding deals, and I am a big fan of these... Basically since the pandemic I think the Nike Blazer has got a revive, they reissued the 1977 style.

A lot of the kids are wearing them, I've got a pair, a couple pairs now and I just ordered a new pair which is the Orange Sail. It's a style. I'm not that into shoes that I understand how all this works, but usually they partner with a designer. Basically the Nike check is bigger than normal, and it's over even the sole of the shoe.

I think it looks pretty cool. When I wear my San Francisco Giants shirt it matches the orange of my shoes, so I'm a big fan of it. It's kind of under the radar right now, so I was able to get it for less than $100, a pair of size 12, men's. I think it's a cool style for something that's not blown up like the Air Jordan Ones. Those are blowing up to a place with unrealistic prices.

Shin: Sounds like it might appreciate the value in StockX in the future.

Brian: Yeah, and going back to the fintech, the original conversation we started with, StockX, I love how fascinating this is about the shoe... Basically street wear culture at this point because they sell clothes now. But to be able to buy something, I don't really care about the value going up or down, I just care about if it's cost effective for me to buy it today because I'm not going to resell shoes.

There's like this whole meme about shoes are made to be worn so I wear my shoes, and if I want another pair of shoes I'll buy another pair of shoes. So anyway, I don't know where I was going with that conversation. But yeah, that's my pick.

Shin: Yeah, and StockX, I think it's a great example of the unbundling of like an eBay, right? I think there's going to be more as the internet economy grows and advances.

Brian: Yeah, and I think specifically in dev tools because you had mentioned Miro being the 1,000 pound gorilla in the room, Eraser is solving a problem for developers in a way that Miro can't do well. Technically they can do, but their focus is so broad that they're getting mass adoption, so now it's just landing and expanding.

So yeah, we'll talk more off air about what I'm working on because I'm in a very similar position, we're hyper focused on one problem that we can execute well on. But with that, thanks so much for the conversation, Shin.

Shin: Thanks for having me on.

Brian: Look forward to hearing more success out of Eraser, and listeners, keep spreading the jam.