Episode 81
· 34:51
What are companies promising to their users? They sort of promise this, like, way of feeling, like, at ease, productive. The important thing, I think, for brands is to figure out what that feeling is and do everything in your power never to break
Jack Bridger:it. Hi, everyone. You're listening to Scaglia DevTools. I'm joined today by Dani Grant, who is the cofounder and CEO of jam.dev, and Jam is a bug reporting tool that developers don't hate. Dani, thank you so much for joining.
Dani Grant:Thanks so much for having me.
Jack Bridger:Could you tell us a bit more about, about Jam? And, actually, before that, I wanna say, like, you've just been completely blowing up on my, like, Twitter feed recently, and I hadn't really come across Jam until fairly recently, and you've just exploded. And so I'm super excited to kind of dig in. But, yeah. Could you tell us a bit about Jam?
Dani Grant:That's so amazing to hear. Jam is like a flight recorder for web application bugs. So when the product manager or the QA tester is, like, checking out staging and something goes wrong, they just click on the Jam extension, and it copies to their clipboard a link they can share with engineers that has everything the engineer needs to debug. It has the last 30 seconds of the DOM playback like a video. It has everything that's in dev tools, like console logs, network requests, the timing waterfall, the stack trace.
Dani Grant:It has all of the session metadata and even things like user IDs. The engineer just opens up this link, and they've got, like, the perfect rerecording of the bug. They don't have to ask any follow-up questions. They don't have to hop on a screen share. The reason why we built this is because when my cofounder and I were product managers at Cloudflare, we we were trying to move really, really fast, and we were trying to ship really cool new things.
Dani Grant:Like, both of us got to be product managers on Cloudflare Workers, on their DNS Resolver 1.1.1. And instead of spending our time, like, you know, unblocking engineers, we were just spending our time chasing down engineers, trying to show them, like, let me show you this is a real problem that users are having, and engineers not being able to reproduce stuff. It's, like, super frustrating for them and, like, waste their whole day. And so you move slower as a team, you do less and which kinda sucks. And so there's no tool that helps the non engineers speak to engineering and, like, show them what's happening in software.
Dani Grant:And so we had to build the tool ourselves, and that is Jam. And so fast forward to today, there are a 100000 people who use Jam. They're at
Jack Bridger:That's more than when I saw your I felt like I saw 80,000, so that was you grosses.
Dani Grant:Oh, yeah. And, yeah, they're they're at, thousands of a r AI startups, but also at, like, the Fortune 1 100. It's like Salesforce and Nike.
Jack Bridger:Wow.
Dani Grant:And, yeah, it's it's been an exciting time.
Jack Bridger:Wow. So people using Salesforce can expect big improvements.
Dani Grant:All the bugs should be fixed. Great.
Jack Bridger:This episode is brought to you by WorkOS. At some point, you're gonna land a big customer and they're gonna ask you for enterprise features. That's where Work OS comes in because they give you these features out of the box. Features like skin provisioning, SAML authentication, and audit logs. They have an easy to use API and they're trusted by big dev tools like Vercel as well as smaller fast growing dev tools like Nock.
Jack Bridger:So if you're looking to cross the enterprise chasm and make yourself enterprise ready, check out WorkOS. We've also done an episode with Michael, the founder of WorkOS, where he shares a lot of tips around crossing the enterprise chasm, landing your 1st enterprise deals, and making sure that you're ready for them. Thanks, WorkOS, for sponsoring the podcast and back to the show. Yeah. One one of the things, like so there's a great demo, that everyone should go see on on the Chrome store, and there's Bob Marley soundtrack in the background.
Jack Bridger:And, I the thing I was amazed by was, like, the yeah. You can actually see the session before, like, the bug occurred even if, like, you weren't planning on like, if you weren't debugging, which I thought was really cool. So, yeah, it feels like things have just exploded recently, from my perspective on the outside. So what do you think is is that how it feels on the inside as well, or have I just been in the random corner of the Internet?
Dani Grant:It's so amazing to hear that. That's really, really cool. You know, like, we live in the coolest time in history because, like, what software means is that one individual or a small group of individuals can have impact to change the world. Like, software brings down the bar to make worldwide impact. And I know that sounds a little cheesy, but I I don't think it's I don't think it's wrong.
Dani Grant:I think that's right. Like, if you think about who made education free to the world, it's YouTube and TikTok. It's like these, like, small and mighty engineering teams at one time. Now no. They're big teams, but, like, they really do change the world.
Dani Grant:And so giving back engineers time where they're not spent, like, trying to communicate with their PMs or, like, chase down debugging details means that they can spend more time building new features and build, like, the future. I mean, it literally means progress can happen faster in the world. So it's really, really cool, and so we really care that we can help more and more people. What's cool about being a founder of a tool that spreads through usage, like, it's, like, product led growth. It's, like, the more you use the tool, the more the tool grows because other people see the tool in, like, Jira tickets and, like, you're sharing bugs with engineers, and the engineers know about it, they ask other PMs to use it, is, like, basically, if we just work on product and trying to make as the most valuable product as possible, more and more people use it.
Dani Grant:And the more people who use it, the faster it grows. And so, it it it has been growing really fast recently, and it it's cool to hear that even from the outside, it feels like an inflection point because on the inside, it feels really exciting.
Jack Bridger:Yeah. And do do you think, there was any, like, big change in how you did things, or was it like a kind of compounding? Or
Dani Grant:Yeah. It's it's so one of the hardest parts of building a start up is right at the very beginning. How do you know if something's working? Because oftentimes things work a little bit. And you're like, is this working a little bit or is it working a lot?
Dani Grant:But when the numbers are small, it's really hard to tell. So, like, our user base doubles every 3 months. And when it doubled from a 1000 users to 2,000 users in 3 months, like, that was it was honestly a little hard to notice. But when it doubled from 10,000 to 20,000, we were like, oh, that's quite interesting. Or 50 to a 100, like like, that's that,
Jack Bridger:you know That's interesting.
Dani Grant:Or, like, okay. Something here is working. And so there was no, like, inflection point, but I think it's just the nature of morality.
Jack Bridger:Yeah. That makes sense. And I know you've spoken, so shout out to DevTools FM. You spoke a lot about, like, kind of how you pivoted 7 times. You you had 7 iterations, and then there was, like, one more was the final form.
Jack Bridger:And I think I'm gonna give a very brief summary so we can people should listen to that episode as well. But you from what I took away from it was that you felt very much like there were you were pushed to do, like, iteration iteration, but then actually, you know what? Can I ask you to explain that? Because I think I've done a terrible job.
Dani Grant:The, the company is almost 4 years old. Like, I started it when I was or I started with my cofounder when I was 26, and now I'm 30. Like, it's been like, that's, like, a huge part of life, you know? Yeah. It's been an amazing journey.
Dani Grant:In the first two years so we started it to solve a problem that we had faced ourselves as product managers. And we really cared about solving it, but we didn't know what a product should look like that solves this, like, very human communication problem. The first thing we did when we started the company is we did 45 user interviews, like, just out of the gate, which is, like, very, like, product manager takes on startup, you know. So, like, stuck to some users. And but what we what we heard on those calls, people were so they they were so frustrated.
Dani Grant:There was so much emotion in these calls that we knew we hit a like, we hit a nerve. And I usually remember, like, one of these early user interviews, the guy tried to sign up his whole team and pay us and schedule an onboarding, and we're like, there's no product yet. Like, we're just at the beginning. One of the earliest user interviews, is this really incredible designer, named Jeff Anders, who was so resonated with the problem that he was like, can I contribute in some way? And he made us our logo and our whole brand colors and fonts and everything Very nice.
Jack Bridger:That
Dani Grant:we still use today. Like, it was just amazing. So we just we thought, like, oh, we figured it out. Like, people are so excited. All we have to do is build and launch, and the rest will be history.
Dani Grant:So we built, and we launched on Product Hunt. And, like, lots of people signed up, and then no one kept using it.
Jack Bridger:Interesting.
Dani Grant:Like, that sucks. Let's try it again. And so we made some changes based on some feedback. We launched again. And again, lots of people signed up and then no one used it.
Dani Grant:And, basically, that was the story of the company for almost 2 years. The 1st 18 months of the company, we just kept launching and launching and launching. We launched 7 different flavors of jam, all different, like, ways to solve this problem around PM to engineer communication and, like, streamlining it around bugs and fixes. And, you know, the the none of them worked. And then suddenly when we shipped the 8th, it did.
Jack Bridger:And, yeah, I was I guess that's gonna be your explainer, but like, what what is not working versus working?
Dani Grant:So think to like, okay, we're a quarter of the way through 2024. You've probably tried a lot of new products in 2024. Like yeah? Yep. How many of them have you stuck with and kept using as part of your weekly workflow?
Jack Bridger:Yeah. A lot less. A lot less.
Dani Grant:Is it even 1?
Jack Bridger:It probably is 1. Yeah. For sure. But, yeah, I think I tried, like, 2 or 3 different video editors this week just for, like, subtitles and yeah. But not stock of any of them.
Jack Bridger:So yeah.
Dani Grant:The bar is so freaking high
Jack Bridger:Yeah. To
Dani Grant:actually get someone to use something new because they have to change their behavior. They're taking a risk. Like, and so the difference between something working and not working is that it's just retention. It's are people checking this out? Is it interesting to them?
Dani Grant:Or it does it become part of their core weekly workflow? And the reason why we knew the 8th version was working is because we just tracked one metric, which was retention. And what we saw was, like, streaks. People were using it every single week, and that felt different.
Jack Bridger:Yeah. That's so interesting. The the guys from digger.dev came on recently, and they kind of basically said the same thing of, like, they would have they would get, like, hackney stuff and people would say, oh, this is really cool. And they'd, like, kick the tires, but no one would ever come back on their early iterations.
Dani Grant:One of the really helpful things as a founder is to sort of break down problems into sub problems because then you can sort of just tackle 1 at a time. Because my startup isn't working is too big of a problem to solve. Right?
Jack Bridger:Yeah.
Dani Grant:But if you're getting sign ups and interest, you know you hit on something that has problem market fit.
Jack Bridger:Mhmm.
Dani Grant:There is a problem in the market and people are willing to go out of their way to try something. So that's great. Right? You solved that problem. Now the only question is can you get to product market fit, where the product solves the problem in such a way that people wanna use it every week.
Dani Grant:And so it's actually really validating that you should keep going when you have problem market fit. Because it means that once you figure out the product, people will like, there are a lot of people who are gonna be coming in to use it. There's this really awesome chart in Lenny's newsletter where he shows kind of the most popular new software company new like, from the last 10 years, most popular software companies, and then how many years it took them to get to product market fit. Like Notion. Love Notion.
Dani Grant:I'm a super fan of Notion. Took them 3 years to get to product market fit. Figma in that chart is, like, the last row to 5. And these are these amazingly, like, these smashing success of companies. But what it shows is that it takes a really long time to figure out a new product that people are willing to adopt and use every week.
Dani Grant:And if founders feel the problem market fit, they should just keep going because the payoff is big if you can give it enough time.
Jack Bridger:Okay. Weird question.
Dani Grant:Great.
Jack Bridger:Sorry. Yeah. It's related. So imagine you knew what you need now, and you had to build, like, the products that was that would get product market fit? Like, could you could you do it in, like, a insanely short amount of time?
Dani Grant:It's it's a really interesting question. Look. The best shortcut to product market fit, if you don't if if you're the type of founder that just doesn't want to, like, iterate through the desert, is you do something that already exists, and you just do it a little bit better. So here's here's an example. Like, linear.
Dani Grant:There's been so many, like, issue trackers. Issue trackers have product market fit. We know that companies use issue trackers, but they did it better. The execution was faster, better designed, more ergonomic. And so that's like a shortcut to product market fit.
Dani Grant:You don't have to invent something new, you just do it better. Like, if you wanted to build a better Salesforce, you don't have to do the winding road to market fit. You're just kind of you just do it. So I think that's one shortcut. But one thing that I've seen now over and over again, it's very humbling for me is, when I meet other founders where their last company is one of the companies that I look up to.
Dani Grant:It's, like, inspiring, smashing success, incredible product, incredible vision, incredible team. And they're starting over, and they have to do the same process that every founder does, which is iterate for years. That's, like, very humbling. And I just think that that's the process. That's that's why it's hard to start a company because the beginning is kind of murky and not that fun.
Jack Bridger:Yeah. Yeah. That that makes sense. But just like getting kind of I don't know if it's philosophical, but, like, in a sense, like, let's say someone comes along, and they're like, alright. I'm gonna start jello dot dev, and I'm gonna just copy you exactly.
Jack Bridger:And feature for feature, just do exactly what you do. I feel like this is it part part of it is, like, your the journey that you went on and and the kind of the story that you tell and that I kind of come to the website, like, trusting that you're obsessed with finding the best way of capturing bugs. And I'm just gonna assume that you're better than jello.dev because they started, like, yesterday, and they just copied you sort of thing. How much of that plays into it?
Dani Grant:A lot of times, especially, like, people with my background, which is, like, product, versus go to market. Like, I come from the product world and now doing the startup thing. So when people come from this world, often we think of companies as their product. Right? And then when you start the company, people, like, tell you this weird thing.
Dani Grant:They tell you, like, company is more than your product. Like, a company is your go to market and your brand and, like, you know, it's your value. It's, like, all these other things. And you're like, yeah, yeah, yeah. But, like, mostly it's black.
Dani Grant:Right?
Jack Bridger:Yeah. Yeah. Yeah. It's features. Cool.
Dani Grant:But, but over time, I've I've seen that it that's actually kind of missing the point where we've seen actual like, we've seen companies that are their products are pretty much similar, but the companies are in sort of different markets. Like, the easy example here is, like, Uber and Lyft. Different types of people use Uber and diff and Lyft or, like, maybe the same, and it's just, whatever, race to the bottom. But or maybe another a better example would be, like, oh, Datadog versus Sentry, where, the product in some of their product categories are very similar, but their go to market and their messaging and their brand is different. So, like, Sentry is the product that an individual developer signs up for, puts in a credit card, and goes.
Dani Grant:Datadog is doing a top down motion. They're gonna be at whatever enterprise conference. They're sending out cold emails. Like, they've just got a different motion. And so oftentimes, like, even even, like, if you copy the product, you might not copy the other things.
Dani Grant:The company is sort of more than anyway, hopefully, that resonates. Yeah.
Jack Bridger:No. No. Exactly. Yeah. Yeah.
Jack Bridger:I feel like it really is. And I I think the founder of Century, I don't know if you saw, but he wrote about brand quite recently and about, like, they've just obsessed over brand and they've invested really heavily in brand and that's, like, their thing, but you don't hear that. I at least didn't hear that much among most of the people I've spoken to, at least.
Dani Grant:They crush it, though. Like, whoever is doing copywriting in, like, the century designs are like, they are amazing. When you wait for the century dashboard to load, it says loading an absurd amount of JavaScript and it's like, you get it.
Jack Bridger:Yeah. Yeah. So kind of branching from that, like, how do you think about the brand of Jam?
Dani Grant:Look, we no developer says, like, you know what I really love about software engineering is when stuff goes wrong and I have to fix it. Like, no developer has ever said, I wish I could spend more of my time fixing bugs.
Jack Bridger:Yeah. I wish I could get more stressful text from my boss on Slack, but, like, saying, wait, I thought we fixed this one.
Dani Grant:I really like it when the company, like, types in the dev Slack channel something is broken. And so it's so the moment that people use our product is not their favorite moment in their jobs. And so our job is to get out of their way, make things fast, make things simple, and keep things light. And that really informs the brand. Like, a lot, similarly, so that's on the engineering side, but also for the person reporting the bug.
Dani Grant:That person was busy doing something else. When they noticed something going wrong and it's kind of a freak out moment, how did it happen? Like, it's if they're busy, like, in that moment, we want them to feel in control, and we want them to feel, like, less cognitive load. They don't have to switch context. And so that also informs the brands.
Dani Grant:Like, keep things easy, simple, dead simple. And so I see the brand as this, like, utility. But beyond that, I also think of brands as, like, what are companies promising to their users? And it's they, like, sort of they sort of promise this, like, way of feeling, like, at ease, productive, like, in control. And and but they but they deliver that brand, they deliver that promise by making you feel that way every single time.
Dani Grant:And if one time they don't make you feel that way, you kind of feel betrayed. It sort of breaks the brand. Like Apple makes me feel high quality. Like, Apple makes me feel like, like, pres like pristine design around nice things.
Jack Bridger:Yeah.
Dani Grant:If one day they deliver an ugly phone or, like, whatever, like, they kinda broke the trust there. And so, the important thing I think for brands, especially for startups like ours, is to figure out what that feeling is and do everything in your power never to break it.
Jack Bridger:And how what kind of things do you do to make sure you never break it?
Dani Grant:The first thing is to identify. Right? Like, have you ever watched a TV show where the first season is perfect and then they ruin it in the second season because they misunderstood what was good about the show?
Jack Bridger:Yeah. Yeah.
Dani Grant:So many times. It's a really funny show and the second season is like the super drama and you're like, you did not always good here. So you have to put it into words. For us, it's like, you know, it's just the words I just used, right? It's like dead simple, easy, fast.
Dani Grant:So once you have the words, then you just evaluate everything you do internally against those words. So for example, when we are looking at a design and it's like, is this ready to be implemented, we ask, could we make this even more dead simple? Could we make this feel easier? Could we make this feel faster? And so that's, like, just the guiding questions and conversation within the company.
Jack Bridger:That makes sense. I think Zeno from resend says, like, similar kind of thing of, like, having that before you put it out there. Do you like, how do you I kind of asked them the same thing, but how do you know when it's like, okay. Yes. This is simple.
Jack Bridger:Like, when it's like this kind of, like, spectrum broad questions. Like, is it simple enough? Or, like, is it
Dani Grant:It's I mean, you don't know until until you see a user use it over a screen share, and then you see them Do
Jack Bridger:they? They get close.
Dani Grant:Okay. Then you know. Yeah. For for
Jack Bridger:the podcast listeners, Danny went very close to her screen as she said that.
Dani Grant:But you can make some guesses. Like, over time we've learned a couple of patterns that are really confusing to users and a couple of patterns that keep things dead simple and easy. Like, for example, something that's really popular right now in in UIs is to do sort of, like, alternatives to standard web components, like making your buttons more square, for example. But that puts, like, cognitive load on the user because they've never seen a button like that or maybe very few times, and so they have to kinda think about it and so it doesn't feel dead simple. And actually, if you just use the standard button that, like, is a material UI, everyone knows where to click.
Dani Grant:Yeah. Or another one is, you know, a lot of pro tools use a lot of icons. And that makes sense because the pro tool, you spend a lot of time learning the tool, you become an expert, and then it's all about, like, speed. And and if you could have all the icons right there, then you have all your functionality right there. But if you're trying to keep things dead simple in a moment of panic when there's a bug, actually probably wanna show people the words, and it's all verbs.
Dani Grant:Right? It's like take a screenshot, share your last minute, and you don't wanna just have the icon. You wanna make it easy for the brain to know they're clicking on the right thing. So, anyways, we've we've, like, identified some patterns that help us, like, just sort of disguise.
Jack Bridger:Yeah. That that that makes sense. So, like, patterns, and then you also combine that with, like, just seeing how people use it. That makes sense. Actually, another question I wanted to ask you is, like, we touched on it at the beginning that it's kind of you've had quite a bit of growth recently.
Jack Bridger:And, how do you decide, like, what to do now that I I'm imagining that you get pulled in more directions than you were previously when you had, you know like, now with a 100000 users, you probably get more pulled in more directions than when you had a 1,000, for instance.
Dani Grant:Okay. It's so interesting. I thought that would be the case, because as a PM at Cloudflare, your users ask you for the whole gamut of things. But in a startup, usually, there's, like, the next logical thing to build, and everyone asks for it at the same time. So every I email the top 100 users from the week before, and I ask them, how's everything going?
Dani Grant:What do you wanna see next?
Jack Bridger:So that that feels like a big thing. So you every week you email the top 100 users?
Dani Grant:Yes.
Jack Bridger:Like, manually or, like, you're just
Dani Grant:No. I have, there's a little bit of manual work here, but there's basically I just, like, have a script to grab those users and then put them into an email tool. And and the email is, like, hey, just checking in, like, how's everything going? What would you like to see? And and then it's me in every response.
Dani Grant:Yeah.
Jack Bridger:Yeah. And people respond.
Dani Grant:Yeah. Which is quite I mean, it's the top 100 users from the week before. Right? So who have just done a bug bash or, like, you know, the like, it's relevant. They're they have something to share.
Dani Grant:Yeah.
Jack Bridger:Wow.
Dani Grant:But the thing that's interesting is most people ask for the same thing. We get, like, the same 2 or 3 feature requests at a time. And then once we do one of them, like it's launched, then it just gets replaced with another thing. There's just always sort of, like, when you're working on a new product until it gets to this, like, place of maturity, there is just an obvious next thing to add that everyone is, like, why don't you have this one thing?
Jack Bridger:Okay. So and and do you spend a lot of is that primarily your time is just kind of making sure that you build those things.
Dani Grant:Product is really, really important. It is it is life or death for a start up. It's also one of those things where it's like, in start ups you work really hard and, you may as well make sure that what you're spending so much time doing and, like, making other sacrifices for is something that'd be valuable to other people. Otherwise, it's kinda seems like a I mean, what a what a waste of time. And so, and so it's just really, really important to us to be close to our users.
Dani Grant:What ends up happening is so, users respond. We have all sorts of user communication channels. All of this information goes into a Slack channel internally called users say the darnest things where then we all, as a team, we, like, discuss it with red, like, how to improve things for users. Sometimes it'll be like, oh my god. That's a quick fix.
Dani Grant:I'm on it. Or, like, yeah. Obviously, that that should be this. Like, let me change that. And sometimes, like, oh, that's a project we should do this soon.
Dani Grant:But that's sort of where, like, feedback becomes action. But because the whole company is involved in, like, all the insightful user feedback that users are, like, so generous in giving us, it it feels quite obvious to everyone. Like, oh, we just get so many requests for blank. Like, obviously, we should just build this next. So it's it's not like corralling sheep or anything.
Dani Grant:It's not like I have to make sure that we build it's like everyone is just on the same page. Like, we are all focused on this.
Jack Bridger:Yeah. Yeah. That makes sense. And most founders probably haven't worked in in product at Cloudflare and, you know, got that background. Like, do you think there's things that work quite well for you that maybe other founders wouldn't know to do, on the product side there?
Dani Grant:Clubflare is such an amazing company. The there are a few things that I think we learned there that, like, I aspire to do. I mean, it's just, like, an incredible organization. Cloudflare is a very audacious company. They always ask could we could we be more audacious?
Dani Grant:Like, Cloudflare made, you know, TLS free for the web for the first time ever. It's like not only did they launch the ability to, like, self serve, get a search, but, like, then they made it free. They're always like, could we be thinking bigger? And so, I mean, I really aspire to that, but I think that's one thing, like, just seeing, like, yeah, really inspired me. And, and maybe if if you've not worked there, like, that's that's one thing I would pass on.
Dani Grant:The other thing that Cloudflare does really, really well is, the in companies, different things are rewarded in at Cloudflare, the the best thing you can do is ship. You can get something out to users, and that is awesome. It's just like velocity. It's all about, like, getting stuff into users' hands, and and I think that that is is really, really important. And and so, we we we try to do that as well.
Dani Grant:Maybe the last thing that, we learned at Cloudflare that is not obvious and now we follow it at Jam is Cloudflare when they built net new greenfield stuff, like Cloudflare Workers. Stuff that, like, they've never done before. There's not really anything like this out there. They build it in a very particular step by step way where the first thing they did is they do is they build a proof of concept that is built to be deleted. But basically having something you can use really quickly is such a great starting point for should we actually build this, what should it be like.
Dani Grant:It answers so many questions. But if you spend too long trying just to get to that initial proof of concept, like, it's it's a lot of wasted time. But if you also build a v one too quickly, then you're gonna have, like, a messy code base as the foundation for a future huge project. And so they do this, like, proof of concept super fast. It's just for everyone internally to try.
Dani Grant:It gives us so much intel, and then it gets discarded. The other thing that's in this, like, step by step is once they get to, like, here's the thing we want to ship, no external person uses it until it's good enough for the Cloudflare internal team. And there's so much iteration and learning that happens in this dog flaring stuff. We do it now at Jam. It changes the game for us because that feedback loop is so fast.
Dani Grant:And if it's not good enough for us, how dare we think it's good enough for someone else. Right? And so it's aligned with our users, it's aligned with how we like to build, it's like fast iteration. So maybe that's also one that, like, we learned there and I think everyone should use.
Jack Bridger:That's amazing. So, firstly, you would build it and then test it internally and then delete it or, like, discard it.
Dani Grant:Yeah. It's yeah. It's it's the duct tape thing. Yeah.
Jack Bridger:Yeah. Yeah. Yeah. And then start from scratch, build it properly, get it working, and used internally Yeah. And then put it out.
Dani Grant:Yep. And then you put it out into beta first, and then you put it out for general access. So there's still also, like, a test for users.
Jack Bridger:That's really cool. That's really, really cool. And is there often where you kind of build these, like, audacious you're doing stuff with AI now, I guess. Right? Like, you you've started doing, like Yeah.
Jack Bridger:Yeah. Woah. We got it in. I got your buzzword in. It's the AI stuff that's you're building, that's pretty new, so I guess you're following you're you're trying to follow the Cloudflare path there.
Dani Grant:I mean, gosh, you just see this stuff, and you're, like, obviously, the future of debugging is AI powered. Like, obviously, we don't want engineers spending all their time fixing bugs. We want engineers building new stuff. Like, how like, basically, it's the most exciting thing happening right now. Right?
Dani Grant:It's like engineers getting more time to be inventive and creative and generative versus, like, you know, like, debugging, crawling through the code base, searching. Like, actually, AI is really good at that. But, like, building the future is, like, totally for humans. And so, that is really, really exciting. 1 of, one of the interesting challenges is, look, we always wanna build trust with our, with our users.
Dani Grant:And part of building trust is things should work every time. Because imagine that you use something, like, as a utility, and one out of 10 times it doesn't work right. You could actually never trust it because you don't know if it's gonna work. Like and so that's kind of the thing with AI right now is so we're do we're following these steps. Right now, we we're doing, like, the the proof of concept.
Dani Grant:Is it good enough for us? We're having AI solve our own bugs. And it's okay, but it doesn't get it right every time, and it means that engineers can't trust it. And the and it's like, so we've created more work for engineers to like check the AI's work, and then like some bugs are easy to fix, and then like actually they're quite easy for humans too. And so so that's the challenging thing.
Dani Grant:And and so we're following the space. We're, like, iterating on prongs every day. We're, like, trying to see if there's, like, smart things we can do around fine tuning. But honestly, I'm just super excited. It's so obvious that in 2 years from now, like, debugging will be 80, 90 percent AI powered.
Jack Bridger:That'll be super cool. That'll be amazing. Danny, that's all we've got time for. Thank you so much for coming. Before you go, have you got one takeaway for founders listening, that you've learned on your on your journey?
Dani Grant:Absolutely. I think one not obvious thing that we've learned that I think every founder should know is that there's old startup advice out there, which is, like, be embarrassed by your first product, ship buggy, duct tape stuff together. And I think that 15 years ago, it was, like, excellent advice, but in the 20 twenties where we have mobile and remote, we all rely on software. Actually, things have to be pretty darn good for people to use it. One of the biggest unlocks for us early on is when we decided to stop shipping bugs and start being really production quality even before, like, product market fit.
Dani Grant:Just as the exercise, like, suddenly it it changed the nature of the user feedback we were getting. We got a lot more clarity. It was awesome. But before we go, I I love I love meeting our users. They're all building such awesome things, and I love hearing their take and their ideas and suggestions.
Dani Grant:So if you're using Jam, and if you're not, go to jam.dev, go give it a try. I'd love to hear from you. I'm Danny at jam.dev.
Jack Bridger:Amazing. Thank you so much, Danny. And thanks to everyone for listening. We'll see you again next week.
Listen to Scaling DevTools using one of many popular podcasting apps or directories.
