[AI] Nick did a hack week and here's what he learned

One of the things that's interesting that you just said is, like, get your ideas to life quickly. I'm sure everybody listening to this at some point has been like, wow, that'd be cool if I could make an app to do that. Or what if I could, like, build something that could just do this for me or build an automation that would just handle this? Like, I'm sure there's been all of these ideas in your head over the years of, like, things that would be cool. What's absolutely insane with AI right now is you can just literally build it in minutes. Like, in minutes, you can create a working prototype of whatever that idea was, like, just by simply typing in words, like a natural language prompt to some of these tools, and it'll give you what you need. It's insane. All right, Nick. So I think for this episode, we should just dive right in. You had quite a fun little week. Worked on something called Hack Week. Right. So basically digging deep into different AI tools and tech and trying to build out different workflows. Let's talk about it, because I know there's a lot of people out there trying to figure out how the heck they can leverage AI more effectively. I think it can feel pretty daunting for a lot of people. Things are changing so rapidly. The landscape's constantly evolving. There's so many use cases, it's just hard to figure out what the heck to do and how can you leverage this stuff? And then plus, companies sometimes don't even let reps do anything. So all of those things combined, it's like, how can reps start to think about leveraging AI or companies even? And, yeah, I think sharing your experience from this Hack Week session would be. Would be fun for everybody.

Cool. Let's dig into that. Definitely. One question I don't want to forget that you mentioned there to come back to is about companies letting you do things because something interesting came out. This. That I think is a good tip.

Okay.

Yeah.

Background. Like what, What. What was Hack Week?

Yeah. So two things. What Hack Week was or is, because I know they're going to be doing more, and two, why I bothered to do this. Right. So the first one is called Hack Weeks. Tom Harmon and Johnny. Johnny, I've forgotten your last name and I'm too lazy to go look for it. So I'll tag you in this later. These two guys have tons of experience in, like, app design, computer architecture, type design. If you look into their backgrounds, you'll see they've got loads of experience in that space working for big companies designing front end stuff, they've decided that they these tools have come to the point now where we can all develop our own apps and automations etc. Relatively easy, easily, but it can be pretty daunting and not know where to start. So they did this hack week Monday to Thursday. Lots of office hours drop ins to talk about how far you've got along, share ideas with a group. And the aim is at the end of Thursday. So you start on Monday, by Thursday you have a demo. And just to put into this everyone's gonna be thinking I ain't got no time for that. Most people on this call didn't have time for it, some of them for work purposes and they've been given the whole week and obviously they had much more time. But I think I had about 30 minutes a day to spare on this because we've got a busy work life as well. And I still managed to get to what I'm I would say is a point where I was pretty pleased. So it's definitely worthwhile. The reason I did it is you may see we just launched a tool for all of our people we work with and potential clients and anyone who needs insights. It's called Insights Succession. It's a tool. Great. And you can you explain it? Harrison, you built it?

Yes. Yeah, yeah. So Insights Succession bio basically all you gotta do is throw in your domain of your company, it's gonna go scrape all the information you can from the website, pull together a target market analysis, different features capabilities that you have, differentiat mini competitive analysis. So identify companies that you could potentially target. We actually partnered with this tool called Gosset AI so we have a direct integration with them to pull some of the companies in and kind of give you a sneak peek preview of those accounts to target. We'll also then pull in specific people, again kind of a little snippet of people that you could target where you could find the contact details, be able to write prospecting emails with one click and then I think lastly pulling together like sales discovery questions that you could use in your actual conversations with prospects. So the idea is simply put in a domain and at the output you get this full report that you can leverage. I would say one thing that's interesting is it'll give you objective insights into what is on your website. Because if your website is surface level and doesn't dig into the details or dig into the weeds enough as to what you do and how you can solve problems for people, that's going to come through because this is purely, you know, using your website as, as a database or, excuse me, as a data source in order to generate these insights. So it was a lot of fun to put together. But yes, that was part of, I think, some of the inspiration. Right?

Yeah. So I think the inspiration came that working with somebody that does, understands this well better than most, and the only compliment you're going to get understands this better than most and does it well and builds these really cool tools internally for ourselves and also now externally, I was like, well, I want in on this if this is something I can achieve. I want to make my ideas come to life because I can't draw for shit. This is like, other than cooking, I have no creative outlet. So this is, was a cool idea. So I went along to this.

Well, I think one of the things that's interesting that you just said is like, get your ideas to life quickly. I'm sure everybody listening to this at some point has been like, wow, that'd be cool if I could make an app to do that. Or what if I could like build something that could just do this for me or build an automation that would just handle this. Like, I'm sure there's been all of these ideas in your head over the years of like, things that would be cool. What's absolutely insane with AI right now is you can just literally build it in minutes. Like in minutes. You can create a working prototype of whatever that idea was like just by simply typing in words like a like natural language prompt to some of these tools and it'll give you what you need. It's insane.

I would say it was much, much easier than I expected to get to something I was happy with, but much harder to get from that point to something I'm really happy with. It's like, move this button an inch to the left. No, not all the fucking buttons, right? You know, there's these levels of infuriation and when you're trying to do something that's quite simplistic, but actually the point of getting to something that was workable was very, very straightforward for me.

Yeah, I think there's. And this is one of the things with these tools is they demo really, really well. It's like, hey, I could put in one prompt, I can get this. Wow, this is insane. With one prompt I can get this crazy like, you know, interface, semi functioning app. But then the real, the way to make it actually like meaningful and valuable is a little bit more of the troubleshooting and going back and forth and, you know, talking to the difference talking to the system and getting it to build and have it function exactly how you want. It's like the fine. The tweaking of it is really, I think where you go from the 90% to like that 99% of what you want from a working thing. And so, yeah, and again, this is only like one thing we're talking about which is like generating like mini micro apps. There's so many other like cool workflows and ideas I'm sure they talked about too. Right.

Yeah. So there was lots of things about not even creating an app, but connecting the apps you're already using. So everyone's heard of like Zapier, that's now got natural language. You can just.

What is. What is Zapier for people who don't know?

So Zapier is essentially think of it as a little bloke in the middle of your apps passing around information. Right. It connects stuff. This is, this is a bloke for Americans means man. So it's somewhat.

I've never heard that description.

Yeah. Is how my brain works. Right. In my head there's a little geezer in there just going, I got this information, I'm going to give it to this guy. But obviously is more automated than that. Yeah. So let's say you got WhatsApp, for example, and every week you get the same kind of information coming to your WhatsApp feed. But it's really annoying and you would like to keep that into a semi organized fashion. You could potentially have WhatsApp and Zapier connect the Google Docs connect to an AI API that would then organize that information, categorize it and put it into your document in a nice template. So every week you just go to your document and see, oh, this came in and it was interesting. And this isn't, you know, like it can do stuff and it's just that Zapier is the connection between. It used to be that you used to have to go and do all of this stuff manually. This app needs this. I'm going to go get that. This app needs this, I'm going to get that. Now they've got a little text box that you type in. I want to connect this with this or this and it starts to build those connections for you. So it's much more streamlined. And I think for me that was the realization that all of these things that used to be, you need to know the programmable language. For a while ago that stopped being true. You don't necessarily need to know HTML or Java or whatever, but you needed to know how to connect them all together and that was another skill. And now it's getting to the point where all you need to know how is to write a reasonably accurate prompt which you can get help with from other AI. Right, actionable prompt, give it to that and then test it. So it's really quite easy to be creative and I was quite shocked at how effective.

Well, I think that's an interesting use case or interesting thing that I think people underestimate is using AI to give you a prompt to then use in another system. So go to ChatGPT and say, Hey, I need you to build me a prompt to do X, Y and Z and it's going to give you a much more detailed prompt than you would come up with and it'll do it in four seconds rather than you trying to think through it and, you know, fill it out and spend a bunch of time trying to format it and get all the details in there and you just take it and you copy and paste it into the other app and then that will run. Let's take a quick break from the podcast to talk about what we have going on over here at Succession. We work with 40 plus life science sales companies to help them find more opportunities and close more deals. We provide a range of services to support your sales growth. So whether it's increasing top of funnel opportunities with done for you lead generation in person or virtual sales training reinforced by personalized training plans, recruiting top sales talent, or just building custom AI workflows to help make your team more efficient focused on selling. So if you're ready to transform your sales strategy, visit succession bio and schedule a call with Nick and I and see how we can support you.

Yeah, so my prompt was I spoke to ChatGPT for about 20 minutes on the train and just brain dumping what I want to do. Even to the point where I was describing a user interface and actually talking with Tom and Johnny who are the experts on this hack week, coming from a novel perspective, was quite interesting because they said they would always draw what the interface is going to look like and upload it. Whereas I was just like I'm just brain dumping and I, I spoke it out. I want this box to be this and I want it in this location on the page or what this box to be this and I want it in this location and it feeds into this box. I want this to be a library where I can upload resources. I want a button to do this for this output. Just kept talking and then at the end I said put it into code. And it did it put it into like a load of code. And I was like, well, that's not what I meant. So I was like, actually, I'm just going to go to this tool called Replit and I'm now going to put it into this box. And I told it where I was going to put it right in the language that is relevant for that box. And they did it and I put it in and got a pretty good working prototype there. So that's how simple it is. Yeah, yeah.

I mean, it's really. It's like a. It really almost comes down to the. It's like the creativity. It's like what. Or not even creativity. It's almost like the use cases, which are. What are the things that you do on a daily, weekly, monthly basis that could potentially be automated or enhanced by using AI? And then just build it, just do it. Just go out and test something. Like grab a tool, doesn't matter which one. Zapier, make N8N replit, bold, lovable, cursor, Windsurf, like whatever. There's a hundred different million tools to do any of this stuff. It doesn't matter. You don't have to have selection difficulty trying to figure it out.

Just.

Just pick one and use it and create something. And I think that's kind of what this hack week forced on you, right? It's like, go create something and then you're going to learn along the way how these things work, how these tools work, how they interact, and then you'll be able to use that information to come up with new and other interesting ideas.

I think for me, more important than the specific tool was the understanding limitations, where they are, what was difficult, what was easy, and kind of freeing my brain from constraints to understand how the constraints of those tools work. Because it's different. Like when we think about doing something, we're going to write a note on a piece of paper, we're going to put it here, we're going to do that. The AI can do it differently and try and learning and understanding that process was really valuable for me because. Yeah, just diving into it with some accountability. Hack weeks. If you want the details, let me know because I thoroughly recommend these guys. Really fun and engaging. Um, and actually I told them at the end. I'm pleased I got into the beginning. Cause it was only a few hundred bucks and actually I think it's going to be a lot more in the future because it was really good.

Yeah, no, that's. That's good. I'm glad you did it. I know you were having a lot of fun along the way and asking a bunch of dumb questions, but it's always fun to, to get all the funny questions from the beginner where you're like asking a question, you're like, what is an API key and how does that API key work?

We talk in this language all of the time, right? The like between us, because you use it. I'm like, well, I know API. I know these terms, I know vibe coding, I know these things conversational level. So I could place them in the right position in a sentence generally. Right? But then when it comes to when I need it and what it actually is, I realized the limit of my knowledge. And I realized that by talking to other people that knew what they were talking about, I was sounding like it. Then they'd ask me a question. And it's like when you first learn, like Spanish or something, you ask someone in a shop, how much is this? And they answer and you're like, I don't know what they said, right? It was literally. It was literally like that for a few days. But no, got around that quickly. The one thing I wanted to touch on that I found really, really interesting, one of the people on my course works for a highly regulated company, right? So they can't go and build an app and use it, but they've got teams that could do that. And what she said was the best thing she found out about this use case was, I've just built on what, built what I want. I've not got to try and describe this. Over 20 billion emails, I've spent a few hours, I've built what I want that does a relatively good purpose. I can now hand in this as a demo model and say, make it within our regulations and constraints. And now she's actually going to get the thing she wants. So she's using it as building demo models for her company. So when you as a rep say, we can't do this, ignore that, build it, show it to your manager about how many hours per day it's going to save you and your team. And if they want it built, they can now get somebody to come in and build it correctly. Or, you know, it may already exist. You know, you may find where it is, but I think you should just go do it. Like find that thing that's a real pain in your ass and causing loads of time, but doesn't really have that much impact, if you know what I mean. Like, you sing it.

I think there's a. I think there's a brilliant use case of, like, you don't have to develop a finished app every time, or you don't have to develop something that's completely. You've finished, especially if you're within an existing organization and you know that you're going to hit roadblocks internally. But you could build the prototype and then share that prototype instead of, to your point, describing this, even verbally trying to describe what you're talking about is going to be hard. Like, you just build it in like 20 minutes and then be like, hey, this is what I want. This is what I'm thinking.

And even down the point of as you're building it, you will realize the flow that you thought you wanted isn't quite what you want. Yes, I actually wanted a different output, but I didn't think that was the output I wanted or a different input. So let's change this. I was trying to build us a. A place where we could store lots of resources because we do lots of training and webinars and presentations and podcasts, and then each time we have to come up with new ideas. So I was trying to build, like, an ideation engine that we would then still write the content because we want to, but would give us sort of starting ideas from things we've done before. And I soon realized that, like, I want. I'm going to need a library. But maybe I don't always want to use that library. I want to use a different resource. So I added a checkbox to unuse, like, do not use library. Or I decided that actually Harrison's going to want to use this, and his tone of voice and his ideas are very different. So we should probably put in a profile where you can upload information about how you want to communicate in your profile. So I added that, you know, but none of that was in my original idea.

Yeah, and I think the same thing goes for the Insights tool that I. That I put together. It's that it started off with this idea of, hey, let me just, like, analyze a website and get, like, a report. And I was like, well, how do I make this useful for people? Well, okay, well, I can make it useful by, I don't know, giving them a list of companies that they might want to target. Well, how do I do that? I was like, oh, cool. Well, I can use Gossip and integrate through their API and do that. And then I was like, okay, well, if I have companies, why shouldn't I have people? And then. So, yeah, we used a couple other different APIs to be able to pull in. Here's some people that you can now go and target. And it's like, well, now I've got companies and people. What would I want to do from there? Oh, I might actually want to like send them a message. And so it's like we have like this 27 page document that we use for, like, how to craft, you know, good emails with all the different frameworks that we use for our lead gen agency. And within that, I was able to basically generate a couple of, you know, simple prospecting messages for people to use. And then I was like, well, what the heck else would they want? Well, maybe once you get on a call, because you now have a message to send to a person at a company and you get them on a call, maybe now you need to know, what should I say on that call? And so we put together discovery framework with a bunch of different questions that you could ask. And so it was sort of this like iterative process of saying, here was this initial idea and then it blossomed. It was like, how do I make this more useful for people? And it'd be the same thing if you even just wanted to use something internally. Like, we've got this internal app that I built for helping to execute a lot of the internal work that we have to do for client delivery stuff. And every day I'm just like, got a new feature here, got a new feature there, like checking off different boxes to like streamline different things. And so now it's like this. I mean, it's not very pretty at the moment, but like, there's lots of little mini apps that solve very specific problems and very specific workflows that we use every single day. And like, it's just systematically making everything we do infinitely more efficient and effective.

Yeah, cool. So and you said API? You mean have I been doing it wrong? Asking people for their appy?

No, I'm joking.

Anyone for their appy?

Yeah, so yeah, no, I think that was great. I'm glad, I'm glad you did it. I know you had a lot of fun doing it, so I think it was a learning for everybody. It's just go do it, get an idea, just go try and build it. And if you're curious or you like just like are stuck on trying to figure out what to do, just send us a message on LinkedIn and we're happy to give our thoughts and help.