Your Team + AI: How to Run a Hackathon That Builds Something

And we managed to merge pretty much every feature into one. So we've got product syncing through, everything from product sync to invoice editing, cost attribution. Your system and your training has helped me a ton with this.

The AI coach will tell you exactly how to handle it. Spreadsheet night is over. The deal moves to approved, the commission gets automatically created, and it automatically recalculates. I actually want this on total. I think they'll love it. On the left is the new generated website, and on the right is their current website.

I personally loved the hackathon. I am so proud and so excited for where we're going.

You guys were amazing.

What's going on, everybody? So we're up for another late-night livestream. For those of you who don't know us, I have the lovely Tara here with me, with me today. She's a colleague of mine at RBP. She is— if you have ever worked with us, you know how indispensable she is in us being successful. I couldn't get through my day without Tara. If she, if she put cliff and jump on my calendar, you'd never hear from me again. She is that— I trust her that much with my calendar, which is that public to-do list that all of you have access to. But today we're going to be talking about how to host a hackathon. And so if you don't know what a hackathon is, it is one of those things where you can shut down the company and it's basically a marathon to hack away at some cool creative problems. And we're going to talk about how we did it because we never did one before. Our CTO is a, you know, he's a principal at IBM and worked for Best Buy and Disney and BP Oil. And he's worked for these huge companies who host these hackathons. And he suggested that we do one. So we were like, great, how do you do that? And he sent us a PDF, and that wasn't quite enough to know exactly everything we needed to do. So then I said, Tara, I think we should do this. And she basically put this whole thing together. And so I asked her to come on the livestream today so she could talk to you guys and show you how we basically came up with this thing from nothing. And if you watched the little pre-roll video, it was an enormous success. Our team loved it. We came up with some really cool solutions to some really challenging problems, things that our customers have been asking for us. from us for literally years. And so, Tara, I'd love for you just to welcome you to the live stream, let you introduce yourself. It's our first time doing kind of a live stream like this, so give her some love in the chat. And, uh, we will— we will— yeah, Tara, let us know who you are and then we'll kind of dive in.

Hi guys, um, thank you for the amazing introduction, Adam. As he mentioned, my name is Tara. I am Adam's EA. I have been with RBP for about almost 3 years now, actually 3 years tomorrow. So that's very exciting. And I am just so excited to be here tonight and talk a little bit more, explain and share all about our hackathon and the entire process that it took for us to Host it and put it on.

Yes, well, that is super cool. And yes, Tara, 3 years is crazy. I just put in the chat for someone to remind me that tomorrow is your 3-year. I'm hoping I'll remember. I think I will. So hopefully I'm not too sick. So Tara, the biggest thing, the biggest thing about the hackathon was that, you know, we— what we really wanted to do is we wanted to get our team ready for being in a startup atmosphere, right? Because we have some pretty big things going on with foundation projects. And so the, the cool thing was is that we gave them total creative freedom. And so I believe, Tara, you had— we like, we called this thing the Forge, and you probably have something that the team built that we could, that we could show them, right?

Um, yes. And actually, those little clips that we showed, um, those were just kind of a little snippet. We'll go a little bit more in depth Later on this evening. But I was actually hoping that we could revisit kind of the point in time where you got this hackathon seed planted into your mind, because I remember it, or I remember when you brought it up very vividly, but I don't, I don't. I think, I think as you mentioned, it was pretty random how it came up.

Yeah, what really happened was, funny enough, I had been doing some digging on some— I've been doing some competitive research on GitHub, and I found a company's GitHub profile, and then I found that their entire CRM, which they're like this huge company, their entire CRM was a public GitHub repo. And I couldn't believe that I found that. So then I went to another repo, like I started searching more names that I knew, and I was amazed to see how many of like in this competitive research, how many people had public repos. And so I went, we were all in a meeting with our CTO, and one of those things that I found was this company that hosted a hackathon, and all of their employees made all these public repos that essentially had live customer code and everything like live, like in open public. Public repositories that you could see. And so I was kind of poking fun at it with my CTO, and he said, oh, we should do a hackathon. And my— and I thought, well, that's kind of a cool idea. And I thought, like, well, and, and we were trying to figure out, like, how would we do this? Like, what— how would we run this hackathon? One of the first things that we always ask at RBP is, like, what problem are you trying to solve? And the main things that I was going for was I have these 2 teams. I have Foundation Projects, this company that's buying roofing companies and deploying AI and robotics. And then I have this consulting firm that helps roofing companies grow. So I have these 2 teams. One of the big goals was to get all the fancy pants over here on this team to start to integrate with the Clydesdales of hardworking, intelligent operators that are on this side. So I wanted to create an opportunity for those 2 teams to meld together. The other thing that I wanted it to be was like an audition for what it was like to work in a startup. Like instead of being coin-operated people that solve tickets and problems for roofing companies, what happens when you have a blank slate and you're going to code something from scratch, right? Based on an idea or a problem that you're trying to solve. And so I really wanted to create that environment. And then the, the, the kind of the third thing was we wanted to create some revenue-generating products. We wanted to say How could we come out with some things, some new revenue-generating ideas at RBP based on the fact that we have all of these AI tools at our disposal? We've tracked every phone call, every email, every ticket, every interaction, every meeting, every webinar, every Loom video in HubSpot. We should be able to mine this for some revenue-generating ideas. So those are the 3 pillars of what we were trying to do. And then that's when I kind of handed it back to you, and I was just like, but how would we do this? And then you came up with the idea for the Forge.

Yes, yes. And when you shared that, I, as you know, um, I didn't truly see your vision. I just kind of heard your idea, and it was like, okay, let's work backwards and identify what you want to happen as your outcome and kind of go from there. And then I think one of the biggest questions that initially came up was, how do we deal with having that dedicated focus time? To do the building. So when we—

Yeah, that was one of the toughest parts. So when you're trying to host a hackathon, this is the problem that we, that we had. And we went back and forth with a number of leaders on both teams because here's the thing, we run a consulting firm. Customers go to our website every day and they're submitting tickets, right? So they're, you know, they built a workflow and they broke it. They have a new employee, they're firing an employee, they need us to get them out of all their systems. They want a new workflow, a dashboard, they're trying to run commissions and they can't— something is not right on their commissions dashboard, or, or we're in the middle of migrating a customer from JobNimbus to Zooper, like all of these different circumstances. How do you just shut down the company? And so we went through a couple options, right? Because we had considered, okay, we'll do Friday, Saturday, Sunday, we'll do a weekend because— and I guess we got to back up a bit. You need a prize. So the prize that we got approved was everybody in the company got a full reset on their Claude Code accounts and $100 of extra usage added to their Claude Code account. And there was a $5,000 grand prize for the winner that was voted by judge. So the voting rubric is something that's really novel and interesting, and I'm excited to share that with you guys in a bit. But there was a rubric for how there was, there was going to be judges. I did not want to be the only judge because I wanted my whole team to win. But, uh, there's no crying in baseball or in hackathons, so we had to make it so that there was a prize winner and there's 3 judges. So we knew we need to shut down the whole company to do this. We thought Friday, Saturday, Sunday, but there was some pushback. Remember, Tara?

Mm-hmm. So yep, the customers were gonna want something, the team was gonna need something, there were gonna be meetings. It was just There was always some problem where some ball was gonna get dropped, and it seemed like we were just going into a no, no, no, or just a negative outcome for any choice that, that we had in front of us.

Yeah, because another option we thought of was, okay, we'll do it 5 days. We'll do Monday, Tuesday, Wednesday, Thursday, Friday, but we'll just do half days. But then we were like, well, then they're not really going to get into the vibe, right? Like they're not going to get into that, that they're not going to get into that flow because they're going to be context switching between work and the hackathon. And so we had to rule that out and we thought, well, if we give them the weekend, but then, you know, like me, I'm that like weekends and holidays are workdays without meetings. So I'm just not human thinking sometimes. Like, I don't think like that. And so then I'm like— so they were like, well, Adam, like, some people have promises to their wives to hang out with their kids, or, you know, stuff like that. Like, you probably can't make people in the last week of the summer, you know, do a hackathon on the weekend and completely abandon their families. So, so like, and like, what happens if one person can dedicate the time but then another person on the team— because that was the thing, we wanted them to collaborate, so we broke the whole company into teams of 2. And that meant like, so how are you going to deal with that? And so we couldn't do the, the, the day, the weekend days, or the splitting it so that we were still open. So Tara asked me a question. She brought me back to that first day, like, what problem are we trying to solve? What do we want to have happen? We want the teams to gel together. We want to come up with some new revenue-generating ideas, and we want to give the team an opportunity to feel like what it's going to be like working in a startup. as these 2 companies merge. And so I said, well, someone's gonna have to suffer, and I guess it's gonna have to be the customers. And that was a really hard decision for me to make, and nobody, nobody could give me any sense of conviction that there was any alternative pathway. So we made the decision. Luckily, we had a month to plan. And so if you're a customer in the room, say hello, give Tara some love for being on the live stream. But if you're a customer, you probably got a series of emails that told you We were going to be shutting down. We shut down Tuesday half day to kick off the hackathon, and then Wednesday, Thursday, Friday we were shut down. Now, as the leader, I definitely knew that if I did it Tuesday, Thursday, Friday, but I made the presentations on Tuesday, I knew I was going to get some discretionary shower time out of these, these folks, and they were going to work on through the weekend anyway. But that's just me being me. So, and yes, they did. And when we definitely saw some PRs get pushed through. So that was how the, the hackathon came to be and how we dealt with the scheduling issues. But what would be another interesting thing is to bring up Tara's PTSD. And so I'd love for Tara, for you to explain where your biggest trauma came into this whole planning process and Absolutely.

So, um, I, I was a teacher for 9 years before I found RBP and met Allison and then Adam and said goodbye teaching, I'm going to RBP. And I would, in preparation for every lesson or any big lesson that I would teach, I would really run through every worst-case scenario. So anytime a question might come up or something's unclear or there was too much transition time or not enough think time, I would anticipate what a student might do so that I could plan in advance for that and prepare for it. And so I came to you with Probably the most ridiculous what-if scenarios in the world. And I think it— you genuinely took it and like you took my initial concerns and you started to like respond. And then I think once it clicked that I was asking these questions because of my background, then you were like, It's going to be fine. Let's trust the process, follow the process, and let's do this. And it was great.

I basically told Tara that she's not dealing with teenagers, she's dealing with adults, because we were worried that we were going to have Sally— that we were going to have Sally the slacker, right, that was going to not join in, or we were going to have Tommy with the backwards hat that bullied his teammates, right? Or we were gonna have, you know, Johnny Do-Good that doesn't, uh, that sucks up. Like, we were gonna have— who was gonna be these teenager equivalents? And I just told— I tried to reassure Tara that these are grown adults. And, and, and we addressed it in the meeting where we said, look, this is what you guys have to do. And so Tara had a really hard time initially thinking, like, was we— like, we went through it, like, do you let them pick each other? Like, do you let them pick teams? And then it's like, you have the same people who always work together who are gonna team up and and go on their teams, and then there's going to be that one loser in the corner that doesn't get picked, right? And so we were like, we, we were like, what do we do there? So where we found a lot of success was Tara went through and kind of did a— I don't know, what was your process? You came up with a— you— the teams you came up with were amazing because we had everybody worked with someone that they don't usually work with, and we also had some last-minute Yes.

So what I actually did was I— another teaching hack— I plugged in and used like a random team generator. So I plugged in all of our teammates and set up, I want partners, and did it randomly. And then from there, I kind of took a look at it and made sure that there was not a mixture or that there weren't Pairings with 2 people from the same team if it could be helped, because I wanted to really merge foundation and RBP. I also wanted people from different roles, so I started random and then I just made sure very small things, rules were in place. And then what I actually did was I gave it to Claudio. With the set of rules to run, and then he did the random picking for me.

So she says Claudio and he— she's talking about my OpenClaw agent that basically helps her and I coordinate a lot of special projects. And so this OpenClaw has every meeting we've ever had, every, every meeting in the company. Like, it's, it's, it's got all of my emails. It's, it's I, we, her and I use it all the time and it really has like a mission of like being very helpful. The other, the, cuz one of the other problems with a hackathon is you're gonna have people on your team who are gonna be very technical, right? Or just like if you're doing this as a roofing company and you wanna do a hackathon, you're gonna have some people who've used Claude a little bit, people who've used Claude a lot, you know, then you're gonna have people who like, I don't like AI, right? I don't have any of those people. They would not survive. At RBP, we're taking that stance. I think I had one person take that stance one time, and, uh, and it didn't— and it didn't— and they eventually just had to come around. I'm hearing that I'm getting some bad audio. Give me one second. It might be— it could be the music. Maybe the music makes it sound better. Give me a yes, no, maybe so on the sound if it's better. Anybody have a— go ahead and let me know if that sounds better, and then it might just be the way it's gonna be. I'll try and talk louder, or maybe it's just that I have like a gruff voice. I don't know. Is that better? Okay, perfect. It's better. All right, it was probably because I had this music thing. I was like, we should have a good vibe, but I screwed that up. Okay, so where was I?

Uh, the vibe is still good. And so we had— we poured all of this planning in advance, in addition, or with, collaboratively with, um, Claudio, other agents. Obviously our full tech stack was used. Um, and then from there, I think we went through a few drafts and versions and started to work through like logistically just all of the really small details. And then when I felt really set, I had the, I guess, presentation or, yeah, slide deck that would introduce it to the team, 'cause we hadn't even introduced it yet to the team at that point. We were only in pure planning mode. What is it gonna look like? How is it gonna work? And then from there, one really big thing that I appreciated was doing a practice round with you. And I asked you to sit there and pretend that you were a teammate that did not know much about the hackathon except that we were having one and you were going to be participating. And let me talk through and explain it. Mm-hmm.

I want you to get your presentation ready and get it into that backstage. But, and I'll tell the— I want to tell you guys a bit, like, not everything went according to plan. So while Tara's loading that up, we went to the nth degree with planning and trying to make sure we had everything, everything ready to go. And what happened is that one teammate, one of the, um, on the investment side on Foundation He's building a health retreat in Costa Rica, right? So he's got a whole property there with a bunch of buildings, and it's this whole like health retreat spa thing. And his— he like hit something with construction, went haywire in Costa Rica, and he couldn't make it. So now suddenly one person was without a teammate, and then another team had a got called in for jury duty, so then she couldn't be there. So then we ended up with a team of 3 that we called Team Spare Parts, and they came out amazing. Like, I, I mean, we might get to show you the project, but Tara did an amazing job of her presentation. And so I'm going to bring it on stage so you guys can see the example of how you would introduce this to your team. So it's one of the main benefits of Running this hackathon is that you can— like, this is like, if you can take something away from this, it's how can you present this to your team? And so this is what Tara built. So we can— yes, we can swim through this.

Absolutely. Okay, so, and really this was kind of the high-level overview that gave the team the most necessary information. Um, that it was just 3-day build. Um, the reason behind it so that everyone knew what their purpose was or what goal that they would try to work towards. And then scheduling especially. Um, want to chat about scheduling?

The scheduling, which part? Like the shutting it— because we did— we talked about shutting down the company, but which part of the scheduling?

Or more of like the proposal time. Did you talk about that yet?

Oh, right. So like kind of the orchestration of it. You're right. So what we did on the first day was we said, okay, so we want you— we didn't give them the scoring rubric right out of the gate. We didn't want them to reverse engineer and go, hey, Claude, help me pick a project that would help me win according to this rubric. We wanted them to start blank slate from— we know you, we want you to create something of economic value. We want you to come together as a team and we want to see you kind of like create something from scratch, right? And so we didn't give them the rubric up front. And so that Tuesday where we half— we shut down for half a day, we gave them— we split them off into their teams. And there was 3 judges: me, Tara, and Jose, the CTO. And what, what we did was bounce— we, we hosted a Zoom room which was like the main Forge Zoom room. Tara created a Forge Slack channel, and so that's where we kept all the updates and the, like, the cool vibes and the fun and the hype. That stayed in the Forge. But then everybody broke off into their own Zoom rooms so that myself, Tara, and Jose could bounce between rooms. Tara mostly stayed in the Forge main Zoom room, and then we bounced around between them to, like, just help give ideas, qualify. You know, you know, just say like, oh, we think you're on the right track with that, or maybe that could be better, or something like that. But we jumped around between the rooms to keep everybody kind of amped up, and that was what we did with the first day. And then everybody's point was to like come basically be able to say, this is what we're working on. We wanted these teams to know how they were going to spend their time. Some teams knew what they were going to do ahead of time because they linked up and kind of Said, hey, this is what we want to do. Other teams, I think, figured it out that— I wouldn't know for sure, but it seemed like everybody came to ideas pretty quickly within that 4-hour block. And so the point was, at the end of the day, they didn't have to make a proposal to get approval, right? But they had an opportunity to say, hey, I think this is my idea. One team actually stopped the screen share. They ran this clandestine operation And wouldn't let me see anything. They wouldn't even show me what they were working on until the 3rd day. So that was kind of how we ran that first, that first day. And then every other day we did the same kind of idea. We would come into the Forge in Zoom, in Slack, kick off a Zoom room, get everybody kind of amped up for the day, and then we'd send them off to go start their own rooms. Tara would stay with Zoom open in the background in the Forge main Zoom room, and then everybody would start their own group Zooms, and myself and Jose would periodically jump between rooms. And one of the hardest parts was the provisioning. Well, I'll let— we'll get to that, but I want Tara to keep on this. But that was kind of how we ran the day.

Yeah. And the proposal, as you mentioned. Oh, speaking of that, it's been 2 weeks since I talked about this.

It's a good reminder. So the other thing that we needed— well, you can go ahead, Tara, you can explain it.

Okay, um, one big thing that we had talked about was getting the pairs and the groups to really focus in and identify and not just like Go and, I don't know, think about or build, go on some random goose chase of building and like go down a rabbit hole that's not a really focused or intentional rabbit hole per se. And so we thought about, again, back to the purpose, um, and Adam, we really looked at kind of specific steps of identifying the problem, who's impacted by that problem, and kind of what it's costing, and this really intentional proposal that required the teams to really plot out their 3 days. And I think that this was so important for the teams to be so successful because it really One, I feel like they bonded during that first planning day, and then 2, it just focused them for the remaining build days. So I like this.

Yeah, and they needed to come to us with what APIs would they need, what tools would they need provisioned, like if they needed a Vercel instance, if they needed an OpenRouter API key, if they needed Like one team needed this API for an SEO tool, another one needed a QuickBooks developer account provisioned. And so that allowed that, that first day— I almost forgot about this. This is the stuff that like you got to think about up front. Like if you do, like if you want to host a hackathon, the reason we're doing this is because there really wasn't something like this that deep dived into what we did. And we also really, we want to relive it Because we want this recording for how— because this was so good that we would run these hackathons quarterly if possible, because it was such an amazing experience. And so we wanted, we wanted to like document the journey to what this was like. And so you don't think about it, but you like, you're gonna— your team's gonna need things. And you have to have a plan for how they're going to get those things. And you're going to do it in a way that's safe and secure. So you don't end up like those other jack-o'-lanterns that I found that had all their all their API keys, environment variables, um, uh, like, like Supabase, uh, uh, Supabase database keys and stuff like this. This is stupid. So this way I was able to keep everything on lock, but they had to come to me to let me know what stuff they would need provisioned so that at 4 o'clock I went and tried to get as much of it ready as possible But then of course their Claude would be like, oh, you need this, you need this. And so I'll talk to you guys later about how I made sure that everybody had the stuff that they needed. Yes, this is so good.

Yes. And so, um, yeah, and we spent a lot more time going over, um, examples, um, Another teacher thing. So I think that it's really important, especially if you're planning your own hackathon or for the future, that you're clearly identifying an end product that would be an example of what you want to see. So we spend time with that. Yeah, of course.

So this is the key when you're trying to build anything, right? It's like you have to ask these questions. Like, I ask these questions in meetings all the time. It's like, what problem are you trying to solve? Because the degree to which someone can explain a problem tells you how well they understand the solution. Because if they can't click, like, like I, like I talk to business owners and I say, well, what problem are you trying to solve? And they'll go on like a 4-minute rant that goes like, well, I want to do this because eventually I think we could do this. And I'm wondering about this. And I mean, AI is probably going to be here in 2 years and then what happens? If Anthropic servers crash, and I think that we could probably build some kind of voice recorder app. I'm like, what, what problem are you trying to solve? What KPI are you trying to solve? Like, and so then I say, well, what would it take to solve that problem, right? What would it take to actually, to actually resolve? Like, what would it take for the problem to be considered solved? And then sometimes there isn't clear— there's like a problem but they don't really know what the solution looks like. So I'm like, well, what do you want to have happen? And sometimes I'll ask, like, what does that mean? Because sometimes people will say something's a problem and it's like subjective happiness. It's not something objective. And I'll say, well, how do you know that that is actually a problem? Or how do you know that that solution would actually fix the objective KPI or goal? And then sometimes they'll say like, what, like, how do you know this is a real problem and we can't just ignore it? Like, so what? Sometimes things are just so what. So this was something to really pre-frame their minds to get specific about what problem they're trying to solve. And I think that this slide is the reason we ended up with so many great projects. Well, it's one of the many reasons, but this Is the foundation, pun intended, upon which that's built.

Yes, I definitely agree.

Yeah. And so this was where we said, okay, in your 4-hour planning process, figure out what you're going to build, right? Like, do you need— like, what— like, like, where are you going to build it? How are you going to build it? What What is the revenue that you think this would generate? What is this competing against? Like, one team I think came out with something that was like the price point that they had was way out of the market, right? And another team came in way low, right? And that's just because you only have 4 hours to do competitive research. But to find out, are you making money or saving money? Because Tara actually brought this up. She said, well, we don't want to eliminate Just— we don't want to eliminate cost savings because a penny saved is a penny earned, whereas you usually have to make a dollar to make 10 cents, right? But you save a dollar, you made a dollar, right? So you add a dollar straight to the bottom line. So we wanted them to be open to both making money, finding things that either make us more money, make our customers more money, but also things that save us money or save our customers money. And so, and they also had to get to a definition of done. That was the third, the second part of the 3 steps.

Yes. Yeah. And I think it helped them visualize and in mind to show, okay, here's what, here's what to build. And it was just so amazing. Oh, and then last but not least, proposals. Part 3, tools and access.

So this is where they had to tell us, and this is— this was the thing that I was the most worried about because like you get— I, like, I remember when I first started vibe coding a few years ago, it was so frustrating when it was like you find out you need something and you're like, oh, I need a— I need a— I need a RAG knowledge base. And now I'm like, Pinecone, Supabase, Vectorize, Graffiti, like which tool do I use? You know, which embeddings model do I use? Like there's so many like things that would pop up that I would get something 80% of the way done, and then I'd have this like administrative task that would like kill the flow, right? And so I didn't want my team to run into that, so I told them like you have to think ahead before you write a single line of code, before you open up your project, make sure you understand what you're gonna need. Right? And so it's like, oh, we need $50 in API credits, we need something for this SEO tool, we need an SDK, we need a database provision, we need an open source thing. So, so this was like— and we kind of gave them some rules like generally yes, you could get these things, and generally no, you're not going to get a $5,000 a month SDK, right? That was such a great step, and I'm really glad you put that in there, Tara.

And I don't think you said no to anyone. I think everyone got their requests granted.

Everybody, everybody got what they asked for.

Yeah, yeah, that was awesome. Um, as mentioned, customer communication— this was big. So I know our team especially, that came to mind when they heard that they will be building for 8 hours, 3 days straight. Is what about the work that the customers that have to be taken care of? And Adam really took care of everything. It was awesome.

Yeah, like when you're running a hackathon, the one thing that I really wanted was I wanted them to have permission to say no to a customer. And that was hard, like, because some of them didn't. Like Dave still came through for a few people, but Dave's pretty good. He's got His own agent harness that he could, you know, send off on a work project and just have it check in with him. But I wanted them to have permission to just disconnect from work, and so I told the team that I would manage the help desk, I would manage the incoming customer requests, I would manage any problems. And thankfully, Codex is pretty incredible. And so if I steal the screen share, share screen. And we'll go here. I used Codex in the hackathon, so I had my Service Work folder. Trying to make sure I don't get cooked here, but I had the Service Work folder that I used to keep in contact with customers, and I and I basically just was like check my email, go into HubSpot, make these make these fixes, and I just kind of like I was using it to like diagnose issues and get. Get these tickets handled. So I basically handled everything from, from Codex so that we didn't have to sit there and be kind of stuck with the customers not getting the answers that they were looking for. And it was, it was a big task because now I had to like jump around between the rooms, but like a hackathon should be high energy. Right? And that was one of the few things that was pretty unanimous in all the places that we— where we could scrape together a little bit of advice, because it's actually surprisingly hard to find good content on how to run a hackathon well. Some of the videos I watched were like 5 years old, you know what I mean? So it had nothing to do with like the AI era. This content is hoping to serve that need that we didn't have. And so I took care of all the customer requests and I occasionally asked for help, but we have good systems. So I was like, hey, I need the username and password for this. We have everything in 1Password suite or LastPass. We have all these places to make sure that I could get access to stuff. We have API keys and system private access tokens and stuff. So I was able to work pretty quickly. It went smoothly. Nobody got too upset. It was a little bit tough coming back to work because it was like All that time away, but that was really good. Sorry, you might have to reshare your thing.

While we're here, it was—

do you guys have any questions? Right, I see some people in the room. I'd love to hear like what you're curious about in terms of running a hackathon. What, what do you think would stop you? Like what questions or curiosities do you have? Maybe your team's like, I don't have a bunch of developers or something like that. We would love to get your comments, whether it's live or afterwards. If you make comments, we respond to all of them, so we'd love to hear about it. But Tara, you can go back to sharing.

Yeah, and another thing to note that we also in the future will plan better for, um, I had planned a trip and I was in Canada and, uh, you really took over for a whole lot. And so I was, yeah, out gallivanting. And in the future, um, we will not plan big events during trip times.

Yeah.

That was— yeah, that was another good lesson, but we learned it and Yeah, it still went on. It did. And another thing, I know you mentioned high energy. One thing that I really enjoyed was that we presented the— we presented the hackathon idea prior— the week before. Yeah, the week before they were set to start. And so they were— they got their team, they got their kind of broad overview of this, and announcing it in advance before the planning time, that initial 4-hour build. I also think that allowed people to kind of drum up in anticipation, and you really saw people's like competitive edge start to come out, and people were, I don't know, Talking trash in the Slack channels that we had, and it created, I think, a fun, just a fun kind of environment for us to be in, and I liked it.

Yeah, it was super fun seeing everybody kind of like rib each other a little bit. It was all in good jest and fun, but like, yeah, to give the team some some some competition, something to kind of like. Something to kind of win. There's a $5,000 prize, so it was kind of a big chunk. It was it was a really inspiring thing to do. Yeah, keep going. Yeah, so this just broke out like how we had the room, how we had the the forge chat. Like and so we we told everybody the hackathon was coming when we had a month to plan, and then we really figured it out two weeks before. We presented this a week before. And so it takes some time. Like, you don't decide, hey, I watched this video last night, let's host a hackathon tomorrow. You want some time to prepare. And so the fact that we had 3 judges also made it so that no one judge had to be the good guy or the bad guy. And then we were able to kind of spend some time together each day to kind of mix it up and kind of get everybody amped up and excited. And I think we even closed out every day. It wasn't really the plan, but we kind of ended up closing out each day as well, which was fun.

Yeah, that was really nice.

The weekend, and then yes, we came down to the demo day. So then everybody, all the teams got a chance to do their presentation. They got 15 minutes, um, and some teams came in right on the dime at 10 minutes, other teams went to like 14, but we, we have 2 hours planned, so everybody got a chance to, to answer a few questions and to present their offers. And they all did so good. And then we deliberated privately afterwards. And then we— so we did the demos on Tuesday, and then we did the awards on Friday. And that gave the judges enough time to go through their rubrics. And if you want to see the rubric, we'll show it to you guys, but we got a few more slides on this thing. So then we picked our teams out, which didn't go as planned because we had the jury duty and the— and then this, this was the part where we kind of just had to say, okay, like, here's another thing, reducing your risk, right? So everything that was produced in the hackathon was work product. It was intellectual property owned by the company. So you can't quit the company and go sell this thing and start your own tech startup. And no stupid security issues. No private tokens left out in the wild. All work must be destructible and revocable— revocable, I don't know how to say that word properly— but we had to make sure that it was there. Um, and then all businesses just had to be either revenue-generating, cost-saving, or revenue-supporting. Think million-dollar ideas. And so there are a few that are million-dollar ideas that came out of this whole thing.

Yes.

I think we have one more page. Yes, so this is just— yeah, this is kind of the one-pager. And then, man, you did such a good job on this.

So thanks.

Now we— So now we've got this thing out in the hands of everybody, we should probably go over the judging rubric. And you did such a good job of, of your— oh no, yeah, you got to talk about this. Yeah, the work environment. So Tara even built a portal for our customers because in telling our customers You created a form for people, for customers, to actually submit an idea so they could actually participate in a way in the hackathon.

Yes, yes, because we— I think you identified like, okay, let's have our customers get involved too. And they can— that was like the buy-in part of, okay, we're really sorry, we're gonna be gone for 3 days, and if you have a problem, hey, here's a great opportunity for you to tell us. So, um, created, yeah, a form that just tracked, and it was the same questions, um, just allowed, I don't know, roofing owners, roofing people to talk about problems that they had and a solution that they might want to see. Um, we didn't get many. I think in the future to change that, I would have promoted it more or asked more people for feedback.

Yeah, I think we needed to put it in our webinars. We could have sent out text message blasts. We sent out lots of emails, but, you know, more or less they were just like, you know, emails that people probably ignored. So we could do a better job next time saying submit ideas for hackathons because we could have talked about it more. But this was great how you created this internal, internal portal Where anybody could go to get their answers, and this is where they submitted their proposals as well.

Yes, correct. And so after the fact, just in case, um, there was— this was like your one-stop shop for everything, um, in addition to kind of that Slack channel. So it just had everything that we, that we covered, um, that would be included in the hackathon, um, along with Yeah, proposal submitting, and then I was actually showing this because I, I thought I put the rubric up here, but I realized I did not, so I'm gonna hand it over to you for the rubric, I think.

Okay, yeah, I'm gonna just find it right now and drop it.

While I Continue talking about all of this great Vercel.

Talk about like, like, like, talk about like, put, put the people in the room at ease who maybe didn't, who don't have a lot of technical background, and how you put this, like, how did you create all this? Because some people in the room might think you're a web designer and a Vibe coder, or like, well, they don't know who you are, so like Walk everybody through like how you came to this.

Hey, um, I, uh, used a whole lot of Claude and yeah, it was pre-switching back to Codex. Okay, so I used a whole lot of Claude. Um, in addition to Claude walked me through how to set up this entire portal. It was phenomenal. Um, helped with that. I was able to— okay, so another thing, I don't even know how to explain this, but, uh, a terminal. How do you explain a terminal in layman's terms?

Um, it's basically like old computers. Like, old computers had a— you, you had a— you just had a black and white screen where you could type into like a screen, and you only had One, it was a command line interface. I think that's probably the easiest way to explain it.

Oh, yeah. But then the terminal, you know, when— because that's how I was pushing everything.

Yeah.

Is— oh, okay. Oh, is that the right terminal that you were describing? Here, let me show you. Okay. I'm— no, not on Claude. On, uh, on, um, on your computer. See, this is how not tech savvy I am, and I turn to Claude. So if anyone is out there and like, how do I even organize this? If I can do it, you can do it. Um, no, so what I was, uh, referencing was pushing updates to get the website live, and then any changes that had to happen, um, they were done in a much more technical way than I have experience or training with.

Um, so you had to submit them with this, right?

Oh yeah, yeah, that terminal. There we go.

Yeah, so fun. Fun fact, fun opportunity for you, Tara, is that if you use Claude, you can get to your terminal. Why is it not dragging? What's going on here? Okay, we're going to share the other screen. If you are in Claude, this is a little lesson for everybody. When you're in Claude, You can open your terminal like this. You don't have to open up this, you can just do it from here.

That is so great to know. See, you learn something new every day. I was with that terminal copying and pasting back and forth to Claude. You learn something new every day. And I'm heavily humbled, and don't— I love working.

You did great. Um, all right, so the hackathon presentation, and you had the— I am not pulling up your rubric, I'm pulling up mine. You made a much better rubric.

Oh, did I?

Yeah, yours have the like, this qualifies as a 10, this one doesn't.

Oh, my anchor grading. Yeah, I know exactly where that is.

Hold on. Anchor there. No. Yeah, there's yours. Yours are way better. I have it here if you don't already. Um.

Okay, yeah, do you want to do yours?

I don't want to. Okay, so basically, Tara, this is the blank version. So how we ranked it was commercial and business value got 30— got 35% weight on the score, and then usability got a 20% weight on the score, innovation 15%, feasibility 10%, Prototype Demo 10 and Collaboration Presentation 10. So we were like, great, okay, so we have a scorecard. But then it was like, well, but how do you rank that? And then Tara came in clutch and basically built this rubric to say this is how you know if something was good, right? It was like, okay, for this— and she said she had a really good piece of advice. She said when she was a teacher You start from the bottom and work your way up, right?

Yes, again, coming in clutch. Teaching background. If you ever need an assistant, find a teacher.

Yes, I can second this. So basically, you start from the bottom. You're like, "Did this doesn't meet this claim? Doesn't meet this claim? Doesn't meet this? Doesn't meet this? Doesn't meet this? Doesn't meet this? Doesn't meet this?" mean this. And then you get your score because I'm like, sometimes you look at it and you go, it's a 10. Well, it didn't have this. Well, it didn't have this. Or like, you just go, well, it didn't have something and it didn't have something, so you go to 8, but really it was actually a 4, right? And so she came out with this rubric, and this helped a ton in the judging process. And what was really cool was at the end, um, I was able to— I know, maybe I can pull up Codex and show In the hackathon folder.

Oh yeah, that dashboard that showed the point spread, it was a great dashboard.

2 things I'm really excited about to show, like where did the hackathon end up after they produced their projects? So I was really excited to see the, the level of GitHub contributions that the team did. We used Greptile for code review. Right. And so we got to see, okay, QBO Invoice was one of the projects. Hackathon Secret was another one, which was a full job costing and invoicing platform. And then Payday for Commissions. And then we had the— this was actually nothing. And then Seven Seas was a sales thing. And then there was an SEO project that was committed locally, so they didn't have it in GitHub yet. But then we were able to see like how many PRs did each, like how many contributions did they make per day. And this is where you saw Saranya went into Saturday and Matthew even went into Sunday. And then, you know, I kind of have this saying that like, yes, GitHub contributions are not a clear marker of quality, but I have this leadership framework that results come from effort, knowledge, and skills. So if you're not, if you want to get the result, you look first at effort. If you're getting the effort, you can add the knowledge. If you're getting the effort and you've added the knowledge, but you're not getting the result, then it's they haven't turned the knowledge into a skill yet, which was kind of a big, was kind of a big thing that I was excited about. But then we also had created this, created this. Which was essentially where the judges aligned. And so it kind of said, okay, this is our scoring range, so this is who won, right? But this gave us like a whole exam of like how everybody performed relative to each other, where everybody's scores lined up, right? You can tell Tara started from the bottom and was better at it, but then also seeing like where was the average and how did the scores kind of play out, right? And so it was really neat to see all these things. And then we also had some places to like review the scores to say like, hey, why did we score things this way? What was like— it just this full analysis of our scores. So being able to use the tool to create these artifacts from the scores really allowed us to see How, like, we processed our thinking around scoring.

Yes, and it was amazing, and it was just so cool. And then the final— I'll let you take it away with the final products. And I know, I think we have Marty and Saranya in the room. I don't know if they want to do a live demo.

I don't think they can come in because they're visitors. But I can show, I believe, Jobline. So the winner, right? Um, and I, again, I was so glad to have 3 judges because I didn't want to pick a winner because honestly, like, right afterwards, me, Tar, and Jose went to a room and I said, you know, we didn't have a scoring rubric yet, and I said, gun to your head, Who would win? And like Matthew kind of ran away with it with his team. But then, and like even afterwards, like I talked to Justin like later that day and he was like, they went before Matthew and they did so, like everybody did so good. Like Marty made this cool video that was really fun and Saranya like had such a clear example of what things were worth. Like Justin had these like customer testimonials in his video. And like Jeanette presented the Super thing so well, and then Dave and Roxanne with their website builder, and like Gary and Adrian had like a really good presentation about revenue, and, and like everybody did so good. But like Justin was like, well, it was good while it lasted. Like kind of everybody in the room kind of said, kind of said, yeah, Matthew took that one. And so I'm really glad that like the rest of the room also appreciated what Matthew did. But this was what they built. They built this thing called Jobline, which we are renaming to Handled, and we've since combined the teams to now turn these— make these— take these projects live. But he had like instant sign-in, connect with HubSpot built into his app. Um, he has like everything that is, that is currently working, everything that's still in testing, things that are coming up next. Like he built in, he had his pricing figured out, he had his security stuff figured out, he has like documentation for like how everything works. Like he did do a really great job. And so essentially his product is making it so that your entire QuickBooks lines up to your HubSpot objects. So it creates bill objects, expense objects, you can do staged and like Um, like staged invoicing. Um, you can do like partial invoices, schedule invoices, trigger them based on deal stages. Like the website needs a lot of work in terms of like seeing how it actually works. We're rebuilding the website now, but they did a really great job of building that. And then I don't know if we have, uh, Matthew set the bar for the next hackathon. Yes, he did. He did a, he did a really good job. Um, and then, yeah, I don't think we can show any of the rest of the ones because they don't have websites yet. That's what everybody's working on now, is building out the presentation for their product. And so the first 2 that are probably going to show up will be, uh, Payday and what's called— being called Handled. And then it'll be Justin's Seven Seas version two program and Dave and Roxanne's Gone Web Fishing product. So, but we're gonna every single one of them, every single one of them was a revenue like was a revenue generating great idea. And so they were excited to have more. We're excited to get all these products into into production and have them available for all of you. So I guess, yeah, what are your like— what is your like closing thoughts, Tara, on the webinar?

Um, I would say, yeah, I would say, um, definitely trust the vision of the hackathon. Plan over plan, um, and Get ready to just be blown away by what your team can create. It was, it was so amazing.

Yeah, the— it was a ton of fun. Well, with that, what I want to do is, um, I will close out with our video, but if anybody has questions, comments, uh, get those into the chat. We'd love to hear what your thoughts on the on the on the hackathon, but I'm going to close out with our little video that Tara and I quickly made coming into this. Can merge pretty much every feature into one. So we've got product syncing through everything from product sync, invoice editing, cost attribution. Your system and your training has helped me a ton with this.

The AI coach will tell you exactly. Spreadsheet night is over. The deal moves to approved, the commission gets automatically created, and it automatically recalculates. I actually want this on Toto. I think they'll love it. On the left is the new generated website, and on the right is their current website.

I personally loved the hackathon. I am so proud and so excited for where we're going.

You guys were amazing.

Yeah, that was so much fun. What do we got here? Great work. Thank you so much, Josie. Appreciate it. And yes, we have some amazing teammates here. Full disclosure, I've worked— I'm working out with our review. I worked directly in SumoQuote, then it was JobNimbus after the acquisition went through. For all amazing companies, but this is the most progressive. Thank you. Thank you so much, Marty. We appreciate you, pal. Well, Tara, thank you so much for joining me on a late-night webinar livestream. Thank you everybody for joining. We will see you guys for another one next Thursday. If you guys have any questions about how to host a hackathon, post it in the comments. Tara and I would love to help anybody else embark on this because this is— this was the kind of team-building event that was a hell of a lot better than doing trust falls in the woods. So yeah, talk to you again next time, everybody. See you later.

Thank you.