I can't imagine a better demo or a better representation of the London AI scene than having Lou from tldraw. Their infinite canvas is wild. Like, it was mind boggling what they did. So I was running events in London before AI demo days. I was working for a consultancy.

I was running the events before that at university. And then when I came to London, they were like, oh, we need someone to run this event. I was like, this wait. You're telling me you're gonna pay for me and my friends to drink some beers and, like, have a chat and maybe learn something? Was like, that sounds really fun.

So I took them up on that. I started organizing events with a consultancy.

What kind of events were these ones?

They started off with serverless events. So very deeply integrated to AWS. We run a lot of events with AWS. And, yeah, that they were really good fun. I met the whole community.

That event's still very strong on Meetup. Mhmm. Like, 5,000 people signed up to that event, which is wild. Yeah. It's crazy.

Like, if I try imagine 5,000 engineers in London Who are

interested in serverless.

Yeah. Who are really interested in serverless. That's wild. Yeah. But I was running that event for a while, and then I also started an AI event in the consultancy.

And then when I left, I like kind of missed it. I missed the in person aspect. I really do enjoy it, and I don't think a huge amount of people do enjoy that type of putting the event together. So I was like, oh, I've got a little USP here. I wanna do it.

Yeah. So I reached out to a few friends when I started at Stack One. And I was like, will you will you help me run this event? Will you like, put up some money? That's always like one of the hardest things about running an event, is I never wanna charge people to turn up.

It's like a dev Yeah. Meet up. But I have like quite expensive tastes. So I do end up having to raise a fair amount of money per event. I don't don't like having pizza.

I think it's very overdone. It's nice to have a nice venue. It's nice to have a big enough space. It's nice to get everything filmed. It's nice to like have photography and like make it feel like a conference.

So I asked around a few places and basically everyone with who I asked was like, yeah, sure. One.

Really?

What is going on? Yeah. Like, this shouldn't be that easy to to get that.

So who's sponsored? The

first one, Single Store sponsor.

Single Store. Okay. Nice.

The database.

Single Store is the one that Nikita from Neon was part of before. Right?

Maybe. Maybe. Okay. They have a surprisingly large presence in like enterprise areas in London. Mhmm.

Yeah. And I've met some of the AEs previously, and one of them had sponsored an event, a serverless, like, party that people were having a year before in AWS Summit, and I got in contact with him. He's a lovely guy, Val, and, yeah, he was just like, yeah, sure. When when do you wanna do it? And so he he helped me out so much that first event.

He like sorted the space, he like paid for everything. Right. They had budget. Wow. He just he did everything, and then I just had to get the people and the and the speakers.

And I was like, this is an arrangement I can do, you know? And we had these amazing, like, falafel buns. They were well, they weren't all falafel. They were like fusion bowel buns. Just remember the food.

Everyone remembers the food.

Fusion bowel.

Fusion bowel buns. So you could have, like, falafel in a bowel bun. You could have, like, there were burgers. There was fried chicken. They were like really rogue, like fusion halloumi ones, halloumi and avocado.

Okay. So wait, how how much did this event cost? The first one?

I think the first one the event space was about £2,000, and maybe like another thousand pound. Maybe like somewhere around £4,000, the first event. So

like 2,000 for the venue. Yeah.

Maybe a bit over 2,000. I think that venue was a little bit overpriced.

Okay.

So in general Yeah. About 4,000 pounds will will run you in an evening venue.

With how many people? And what was the space like

as well? This was about 50 people. Okay. So it was much smaller than everything we've done since. And the space was okay.

It was okay. We could have done better. It was a bit dungeon y, we could have done better. Okay. And then the next one, I was like, we can go a little bit bigger.

We can go bigger. We can go bigger. And so what was the next one exactly? The next one we did at AWS. So the next one I was actually I was a friend of mine offered me a venue inside AWS.

Wow. Cool. And we we we ran it was way too large. Right. Like, it was was that the next one?

I'm getting too confused. We had one in AWS and we had one in Coher, the model company.

Oh, they have an office in

London? Yeah. Do.

Yeah. Really?

Yeah. And we had the guy who invented Rag. He coined the phrase Rag, came

and spoke. And I was like, holy shit. This is wild. They're in London. The the guy

that Yeah. Yeah. Yeah. Yeah. He he came out at UCL.

Yeah. Wow. Yeah. Yeah. Really cool.

I think the eighty first one was next, and then we had Cahir. And at that point, by then, I'd worked out that the people who really like events were VCs. And so VCs love sponsoring things like this.

Oh, sorry. They like to sponsor they they don't

Yeah. They like events, and they like sponsoring events.

Okay.

And it's good for their image, it's great for hiring for portfolio companies. Mhmm. It's good all around, like they get to meet interesting people

and if I good at running events.

Yes. They're they have normally teams of people who this is their job. Yeah. And if I I could do something and it doesn't have to, like, disrupt their team, and they

can just pay me a

little bit of pay the event a little bit of cash to do it, then it normally works quite well for them. And they seem they seem relatively stoked on it. Like, I mean, that they're the things they're getting out of it. Hiring, meeting cool people. Like, they always wanna add value to their portfolio companies if they can be like, we have a demo slot at this, like, the most exclusive AI event in London.

Here, you can demo. Then, like, that's great for their portfolio companies. So we'd kind of worked out the VCs, like, sponsoring. Oh, I have some really funny stories about the Cahir one. I mean, you gotta

tell us.

Yeah. I've gotta tell them actually. Yeah. Go on. I'll tell you.

So so we were all set. We had all the speakers. We had a very we had Cherry VC we're gonna sponsor. It's gonna be amazing. This one was the only one we've ever had pizza before, which was not my choice, but, you know, amazing amazing office space.

Like, we really got the vibe right. Like, it was so good. Everyone everyone was, like, crowded round, but the setup of it was such a pain. So I got a message from their office manager the day before the event being like, just checking. You are bringing your own chairs.

Right?

Oh, no.

And I was like, what do you mean? She's like, well, we don't have chairs. I'm like, it's an office. What office? So this is everyone stand up?

What office doesn't have chairs? And so I was just like kind of flummoxed. Was like very confused. And so I got I got on my bike, I cycled over, and I knocked on the door and was like, can you let me in? I wanna see I wanna go count your chairs.

Yeah. Yeah.

And she was like, yeah, we don't have chairs. And the office man, the office was so beautiful, like, they do such a good job there to keep it really nice. And we just counted chairs. Like, we just counted how many chairs there were, and I was like, we have 50. It's gonna be fine.

There's only

And

these 80 people here.

Yeah. But these were the kind of like chairs people use at their chairs.

No. No. No. This was not those. So I didn't wanna use people's chairs because I assumed they were their chairs.

Did

have chairs.

This was like poufs. This was like settees. This was like kitchen chairs. This was loads of stuff. Yeah.

Yeah. Anyway, there's there's been some really funny ones like that. I think they just hadn't, like there's a big misconception. Like, they they were someone had offered us the space that wasn't the same person. And then most of running events is, like, communicating to people what you're doing.

Yes. I'm gonna be in your office in the evening. Please don't make me sign an NDA. Please don't make me have to, like, fill out loads of forms for everyone I wanna bring in, and, yeah, please can I have it for free? Yeah.

It's like all of those things that are not very second nature to a company, and especially if you've never worked in a developer first company, which a lot of people haven't, then you might have never come across this random, like, person just turning up asking to use your office and being told from, like, higher up that that was a fine thing

to Yeah. Yeah.

Like, literally, the way I got cohere was I DM'd one of the founders on Twitter.

Mhmm.

I was like, it was Ivan. And I was like, Ivan, can I borrow your office in London for an event? And like, I am not a famous person on Twitter. And he replied and was like, yeah, sure. Yeah.

What what do you wanna host? Yeah. I was like, I don't wanna host an AI demo event. He's like, yeah, sure.

That's really cool.

And he called me the next day. And we we had a chat. Like, he literally spent his time to like just be like, yeah, it's fine.

Scaling DevTools is sponsored by WorkOS. At some point, you'll land a customer who needs enterprise features like audit trails, SSO, role based access control. You could spend ages tearing your hair out, building these things yourself, or you could use WorkOS. Let's hear from Uphpal from digger.dev.

I can speak for open source companies because I think that's where I have the most experience personally. If you're open source and you're doing enterprise first, the minute you think about monetization is when you should think about WorkOS. How it's designed is that you can start as early as day zero. But for us, it wasn't day zero. It was closer to when we first started monetizing because we didn't have a sign up at all.

People could just anonymously use our tool. To be honest, if we do that again, I think we'd think about that on day zero, be honest. Because, like, we should have done it on day zero ideally. Anonymous usage should be permitted, but you should know who's using your tool. It should be opt in a 100%.

But it'd be great to have auth from day zero. So, yeah, that's what I think.

Thanks Work OS. Back to the episode. We were talking before and I I think the reason that you actually messaged me about we should do an episode is I think you were listening to the episode on hackathons. Yeah. And you were like, oh, I wanna talk about events.

And also, don't really agree that hack funds are such a good use of time resources for devtors.

Time resources or money. Yeah. Yeah. Which I

have to be persuaded on, by the way.

Yeah. Because

I I feel like I feel like I'm in the in the pro hackathon.

So hackathons are great for the people who go to them. They're amazing. They're really good for your CV if you do well. If you don't do well, like, you've learned something new, potentially, you've experienced something away from your job. If you're a student, you've, like, seen how professionals deal with experiences, like, how they demo, how they well, I mean, they're essentially a pitch a lot of the time, how they build something really quickly, the technologies they use, like, is good if you're an attendee.

Didn't you win a hackathon Yeah. With the government? Yeah. With Sunil. With

Sunil and Who are now

gonna be working?

Yeah. Yeah. Yeah. Exactly. So in our in our DevTools chat, when Sarah wrote, do you find hackathons what did she say?

Do do you think you get a return on

Yeah. Hackathons.

Was about going to them though. Was like, did you, like, meet a co founder? Did you raise money? Did you find a job? Or or nothing?

Chill out Sarah. I think it was the last option.

Sarah Drinkwater. Yeah. Where she wrote. Wrote. Shout out Sarah.

Shout out to Sarah. And I wrote found a job, because I think that was probably the yeah. I think in that moment, I was working with Sunil and Thomas, I was like, these guys are really cool people to work with. And in the future, in some in some point in time, in some alternate life cycle, I would I would like to work with them again. And that's the beauty of hackathon, is you get to like see other people.

But yeah, great as an attendee. As as an organizer, I can see why VCs do it because they it's everyone's struggling for early stage deal flow, and you just immediately get early stage deal flow. So that's really good. I can't see why I can see why open source companies do that a little bit because they have that community focus that they're trying to foster, and you get really good vibes from a hackathon most of the time. But I can't see why any other company organizes a hackathon.

And that's I mean, we were talking before this. I was like, the products that you tend to be told to use at a hackathon are normally quite shit. Interesting. Yeah.

Go on. Tell us about that.

Well, like, everything you're told to use a product because they've sponsored the hackathon. There's a reason why they sponsor the hackathon. It's because they either wanna get, like, free engage they wanna get engagement, they wanna get some user feedback, which probably means, like, the product is not quite ready. If it was ready, they wouldn't be sponsoring the hackathon. They'd just be giving prizes at the end.

They wouldn't be, like, telling you to use the thing. I don't know, like, the amount of times I've been to a hackathon or seen people at a hackathon and they get to the end and they get to the demo and their API keys run out of credits or like like this happened. We ran yeah. It's gonna sound like I'm really hypocritical, because we actually ran a hackathon last week, and I think there was good ROI for that. But the day long hackathons, the week long hackathons for individual companies Your

one was an evening.

Yeah. Mine was an evening. Two hours. Yeah. £5,000

or dollars. Yeah.

Like a silly amount of money, prize money. But we could do that because it's only an evening. So luckily, we got the event space for free. We could just put all the money into, like, giving it to the attendees. And so I think you end up with, like, quite a fun experience for everyone involved, and we got some free product feedback that none of the attendees really minded because for them, like, so many of them walked away with actual money that it was like fine.

So I think that was I think that was fine. But if it if we just like scale that up and imagine that evening hackathon was actually a day hackathon, or a weekend hackathon. Yeah. Suddenly, we have to pay for venue. Like, if it's a day long hackathon, it might be £20,000 in venue, maybe more.

Week long hackathon or a week a weekend long hackathon, you can obviously times it by two. Remote hackathons, I don't think you get any of the benefits of in person hackathons.

Because you're not even seeing people Yeah.

Struggle. So I think you just like cancel that all out. Like remote hackathons is is just such a low bar

Mhmm.

And you just haven't met it if you're running a remote hackathon. And maybe people maybe people find some good go to market there, but I just don't see it. But, yeah, if you're scaling this up to a whole day, your costs are just so astronomically high that if your ROI is just, I'm going to hire someone from this, then maybe you can accept those costs. If your ROI is I'm gonna get, like, brand recognition and all of this stuff, There's probably better ways of doing that.

Yeah. So how how much less how much would a hackathon cost typically do you think? Just for anyone that hasn't run one?

Well, like your day just your venue expense for a day will be about £20,000.

And you have to have it potentially for you might have it for the whole weekend.

And if you're having it for the whole weekend yet, it's more. Yeah. And there are some spaces that were specifically for this type of gig and maybe you got them a little bit cheaper, but most of them are going out of business because, like, they can't sustain rent in Central London, like, for this type of thing because no one should be paying for this type of thing. It feels it feels so wrong. Like, let me give you another reason.

Because I feel like we've been a bit light on the reasons. A bit strong on emotions.

Yeah. Well, because I would throw it back to a lot of the ones I've been to have been held at VC's offices. And the startup usually is like the VC that invests in the startup. Startup runs a hackathon. VC provides the office for free.

So And I know Ratecast

did one like Atomic. Right? Yeah. And that was probably great for them. They just released this thing called extensions.

Mhmm. Extensions are open source. Anyone can code an extension.

Mhmm.

And that was actually the model that we went for disco, for the our disco hackathon. It's like, we release these integrations, they're open source, anyone can code one. We like specifically made it possible to code them with AI. So it's very, like, easy to get started, quick to do, and you're just you're limited by some some imagination potentially. That that's it.

And so that's a really good candidate for a hackathon. And the Raycast one also. They're like a fun brand with cool branding. There was some really good swag, like Mhmm. They'll have gained a lot of brand loyalty in London from that, and they'll have gained a lot of really good extensions.

Mhmm. Because, I don't know, 50 people sat down for a day and built an extension. Mhmm. Like, how much would they have to pay to get 50 people, like, fifty days worth

of building? Just throwing it, like, devil's advocate because you're just kind of saying, like, that it's not worth it for that. Like, you're like No.

For them, that's worth it. For them, that's worth it.

But who is

it not

who is it not worth for?

If you have a closed source product that you're running a hackathon to get people to use your product, and you're maybe a little bit early in the product life cycle, you're spending a huge amount of money, and no one can see why your product's breaking when it breaks. Even if you're, like, slightly larger and you're running a hackathon, and you're, like, one of the prizes, you're gonna end up spending if it's, like, API credits and you're giving away, like, LLM usage, you're gonna spend a huge amount of money per person coming to the hackathon, even forget the costs of like the venue. The best hackathon yeah. The a 16 z one that I met that I worked with Sunil and Thomas on, that was amazing. Like, as for an as an attendee, that was incredible.

But that, like, I I I can't believe that did ROI for a 16 c. I mean, they left London, six months later.

They left London?

Yeah. They closed down their London office. Really? Yeah. Wow.

I mean, that can't there's not a cause and effect there, but if they'd have been doing obscenely well, they probably wouldn't have left. Actually, we have some amazing stories from that one as well. So we made this parliament parliament wow, like the easiest way to see what was going on in parliament, and we just

Oh, cool.

We just summarized a bunch of video, basically. And we did a lot of indexing of the parliament bills going back really long time. I think till the eighties, I think. It's basically all the ones we could get from Hansard, which is the government database online. Because we won this Build Back Britain one, which was like the government track, they actually took us to the houses of parliament.

So I ended up going and, like, having this, like, mixer at the Houses of Parliament. We were, like, paraded around. Really?

I

was, like, yeah, look, these young entrepreneurs are, like, building back Britain and building all of the infrastructure for Parliament, and we were just there, like, this is awesome. And I really wanna do some of this, but we have day jobs. We're probably not your right your right demographic. And that is it. Like, a lot of people who go to hackathons have no intention of quitting their day jobs to go and work on the hackathon.

So even for VCs, right, like, early stage founders might, if they have a good idea, might not have a lot of time to go to hackathon. So a lot of the time you get career developers who work at big established companies going to hackathons, and they're not about to quit their day job to go and work on this idea that they that they bashed out at a hackathon. Like, the people yeah. I don't know. Like, I just feel so so misaligned.

Was there an in person event, like a like a like a demo day, where people actually bring the startups they're working on, and they are ready made companies to invest in. Like, for a VC, that's to me, if I was eyeing up paying £25,000 in sponsorship for a hackathon or paying £4,000 sponsorship for an evening event, and being like, I'm gonna meet a 120 people either way on either of these things. In one of them, they're gonna be hyper stressed, slightly sweaty, and they're gonna be trying to smash out an idea for a share of some prize money. Or in the other one, they're gonna be super chilled out over a couple of beers. We can extend the conversation.

Like, we can, like, really get into it, and I'll have seen them pitch for free. I I just don't understand how you would say, oh, yeah. The hackathon's better ROI there. For a VC.

Yeah. So tell us about the format of AI demo days for those that don't know about it.

So previously when I was working in consulting, we would do maybe two talks per event. And those talks would be twenty five to thirty five minutes, long format talks. And I'm honest, I I just like, I'm part of, like, maybe not quite a TikTok generation, but I I was getting quite bored a lot of the time.

Yeah.

And so I really wanted something shorter, snappier. So when I had my own control of these events and I wanted more startups to demo, I was like, well, why don't we halve the time? So instead of thirty minutes target, it's now fifteen minutes target. And actually, it's ten minutes because we'll leave a few minutes of questions at the end. So, yeah, they're normally five ten minute talks, ten to fifteen minute talks, and we go quick fire.

There's no breaks. There's no time for people to like get up and leave halfway through. Otherwise, the last person feels like they got cheated, you know. Like if you have two talks and people leave halfway before before the last talk, then the last person's like, what? Yeah.

You called me here to like speak to three people? So, yeah, we just got all the way through. The event has, like, good networking before, good networking time afterwards. You make the vibe feel like a conference or not even a conference. It's like a meetup of friends.

Like, I started it because a bunch of my friends moved to San Francisco. I was working on this project called Quiver Mhmm. Which went to YC. And then I was just like, like, why did they have to go to YC? Like like, why?

Why did they have to go to Y Combine? Why did they have to move San Francisco? And there there are many reasons why they moved, none of which I'm gonna fix, like, why they moved out of Europe. But I thought the smallest thing I could do would be to have a space where I could see what some of my other friends or people maybe I can make some new friends who are also doing similar things. And events are really like, in person evening events are a really good way to do that because you meet one startup who has this awesome cool demo, and they've got 10 friends, all of which are doing really cool stuff.

And then you just you just keep ticking them off. Mhmm. And you you work out what you like, you work out what you don't like. A lot of the time, I feel a bit like a VC and I have to be careful that I'm not just having like DevTool founders. Because I've like quite a bias.

Right? Like Yeah. I work on DevTools. So I need to try and make sure that I like get out of my niche every now and again. Appeal to a broader audience.

Yeah. It's hard. What would you say you do like if well, okay. Let me put it this way. If I was saying to you, I I wanna start running a kind of AI event in London, another one.

And besides the fact you might be like, oh, then why? You don't need to.

No. No.

You should. Okay. I should.

Okay. You definitely should.

So I should.

This podcast should have a in person event. People are gonna come because it looks fun. Mhmm. So it never looks fun if you're, like, begging people to come. Does that make sense?

Yeah. Yeah. Yeah.

Yeah. So just don't worry about people coming in general. Like, will come if you do if you think about it. It cool. Yeah.

Play it cool.

True to mean. Keep them keen.

Yeah. You know? Kind of. Not too mean, otherwise, you might just be mean. Yeah.

Get the best demos you can get. Get the best demos you can get. I think we were really so Lou from Teal Draw opened our demo series, open demo days. And I was like, I can't imagine a better demo or a better representation of the London AI scene than having Lou from Teal Draw demo the crazy magical autocomplete teach computer stuff they do over at Teal Draw, like their infinite canvas. It's wild.

And they they this was in that moment where they were really, like, pushing some of the funky demos that they were doing, and, like, it was mind boggling what they what they did, what they did. And so I think that really set the scene. Like, the people who came to that first event, most of them have been to every single one.

Interesting. Yeah. So cool.

We had 50 people signed up. About 50 people came because they were pretty much just friends of mine, you know, like, invited my friends. I didn't have a huge network. And they've all been back. They're like the core group now.

Yeah. I feel like that's quite important. It's like having a

Yeah. I think if Demo Days was a startup, it it hit product market fit like right at the beginning then. Like, everyone came back. Yeah. And that just shows that we need more events like this in London.

Mhmm. When we ran an event in San Francisco, we had some amazing demos. Like, there was the reason why we went was the head of developer relations at OpenAI developer experience, Roman said he would demo and I was like, this is too much of an opportunity to miss. My brother was going was working at a YC company, so I wanted to go visit him. And we managed to put together at very last minute, like one of the coolest demos, demo days I've ever seen.

Like, it was awesome. We had artificial societies like

Oh, yeah.

They're based in London. Yeah. But they were going through YC as well, so they demoed at the same time.

That's cool. That's the one where you kind of get an idea of how your social media post will do. Right?

Yeah. You can put out like a draft post and it will like put it through your whole LinkedIn network or your Twitter network and it will tell you like, maybe you could try this variation or this variation, make this hook a little bit stronger or something. Mhmm. I'm really bad at writing social posts.

Mhmm. Do you use that?

I used it initially. I used it initially. I need to actually start paying for it. It's not something that I've like prioritized.

Which do you think people should?

I I think if you probably yeah. I think if you want to grow an audience, you you should definitely try it. Yeah. Mhmm. I know initially it was very tuned for James, like the the creator.

It worked really well for him, and you can see by his social posts, they're amazing. Every single one of them are, like, really cool at the end. I'm like, why am I reading this? And it's like, I know your story, James, but it is incredible, and he's, like, really inspiring. And I'm really excited for them to roll that out for everyone, like, yeah.

We can talk a lot more about AS. They're they're super cool. Okay. So we're in San Francisco. Amazing demos lined up.

Five really cool demos. We had the creator of Docker as well on his new thing, like, this Dagger. And I was like, oh my god. We've got like some of my, like, absolute heroes in technology demoing. And there were five other demo nights happening on the same night in San Francisco.

Really? I was like, no one's gonna turn up.

Wow.

We've got five other demos. So yeah, London needs way more stuff. Yeah. It needs like

But people did turn up. Right?

People did turn We were full house.

Yeah. Yeah. We were full house. Yeah.

We were really yeah. I I wonder I was wondering up in that moment, like, which other demo night got, like, I don't know, 20% return or 20%

Yeah. Maybe people

Yeah. Maybe there's enough fuel. Which is weird because there's not that many fuel in San Francisco. Think there's seven seven hundred thousand people in SF Mhmm. Compared to what?

10,000,000 people in London?

Yeah. That's crazy. So you were saying though that you want to create an event that people actually like looks fun.

Yeah. Right? Create an event that looks fun. We I get a lot of help with the branding. So Yeah.

All of the branding is done by the Stack One designer, Morgan. Mhmm. So he keeps a very nice brand feel all the way through. Yeah. We actually just went through a branding refresh, which was really cool.

He just out of blue was like, guys, I did need designs for you. Really? Sick. Yeah. That's awesome.

So cool. And so, for me, like, I was showing him how I was doing the designs before, basically in Google Sheet in Google Slides and like screenshotting stuff.

Yeah. And

he was like, man, you're making me feel a bit ill. So he wrapped up Figma and just like did it for me. Yeah. And now I can use Figma because he taught me. And so I can like make them myself,

which is really cool. Frames Can you though?

No. I can change I can change words where he's already written words. Yeah. So And maybe I can put a new image in and lower the saturation. Yeah.

I can do that. Yeah. So I would I consider myself somewhere between intermediate and advanced. Okay. I'm kidding.

I've yeah. I I can do just what I need to do to to to get the stuff out the door. But it works really well. And yeah, so how do you do it? Make it feel fun.

Like, I mean, Damien's done a really good thing with AI engineer. They were cool in the they started with a brand. So Tinkerers also has a worldwide brand. Anyone who's coming to London, they're probably if they're from Canada or they're from San Francisco or even they're from Singapore, they're gonna have heard of AI Tinkerers. And there's a bunch of other places now that they have But I think those are the big ones.

And so they'll come to London, they'll be like, I'll go to the Tinkerers event that's on. Yeah. I mean, we've done quite a lot of the big tech companies. Yep. Previously, I've been to Meta, the office that now closed down in Ruffman Square.

That was wild. All the lights, like, were controlled by this bunker underneath the main room, where there are like four guys sitting in there with joysticks. Really? Yeah. I was like, this is the most amount of effort that's ever been put on for my headband.

Yeah. That's crazy.

So you got this awesome joystick guys.

Yeah. Four joystick guys. It was literally wild. There were, I don't know, maybe 50 lights.

Yeah.

All being independently operated. Wow. And then they made a video for me afterwards fully, like They they made it.

Yeah. That's

My contact at Meta had to get so many sign offs to get it sent out, but it was so cool. So yeah, we got this amazing menu because you've gone to Big Tech and you've asked for it in the evening and they are they're happy to give it to you. It's really good for their brand image as we talked about earlier. It's good for their hiring as well. Every team at that moment was hiring, so it was great.

You've got a little bit of extra cash. You always need some extra cash. Like, this is the way that events fall apart. It's like, you run out of money or there's a surprise bill. There's always a surprise bill.

Like, probably at one in three events, there is like some surprise where someone drops out, a sponsor drops out, something changes, and you have to pay a thousand pounds. Yeah. Your videographer decides he can't do it, and now You have to he has a friend who has a who has some time, but they want more money. Yeah. Oh, okay.

So you always have to raise some money for an event, even if you're like, we're just gonna pay £50 for some pizzas. You can't. You you need to have some actual cash Yeah. For it to be sustainable in the long term. Yeah.

And so that's why every event of ours has been sponsored by a VC since the first one.

Mhmm.

Not just a VC. We have had some external companies sponsor as well. Some of, like, really good London tech companies have sponsored. Yeah. You need to have some safety cash.

It's so sad when events stop. Yeah. When, like, something that's been going for ten years or less, like, eight years or something, and then just one day they're like, yeah. We couldn't do it. Mhmm.

Why not? Oh, well, the last one I got burned about £500 because we had no one to pay for

food or Yeah.

Those problems get bigger as you get bigger. So if you start running an event with a 150 people, suddenly your drinks cost a lot of money.

Do you, Matt, do you think if like a DevTool wants to just run an event though, do you think that's worth it?

Like just for them?

Yeah. So like if they were like, okay, they they were gonna start AI Demo Days. Let's say like Coher was gonna start Yeah. AI Demo Days Yeah. In another time, you know, in another world.

It's hard for them. Like, it's hard for them because everyone wants to see like a like a everyone wants to see their brand image, their ROI, like everyone wants to have their stamp on the event, and the best events have no one's stamp. They're like, they are actually a community thing. Mhmm. Who arrives feels like they're part of

Yeah.

The event. They're just like coming to see their friends demo something cool. I think events that are more like lectures can be good from a company. So Anthropic just put a bunch of events, they they recorded them, put them on YouTube about how they use Claude code, and those, they're amazing. I've watched a couple of them.

But they are lectures. Mhmm. Like, to me, they're not a community event. They're like Mhmm. Anthropic, slightly pitching, slightly giving some good secret sauce about how to use their product.

So that would be a way of running an event. I think they get away with a lot because they're very hypey and people want to be close to them. But that would be a way of doing it.

What about the extreme like you get like AWS re:Invent or something?

Reinvent's a sales pitch for everyone involved. So re:Invent is like a completely different kettle of fish. It's like most companies that go to re:Invent pay a ridiculous amount of money to go to re:Invent.

But like if you're running it there's still kind of like AWS brand, but it's not just all about Yeah. My understanding is

So like the cool one that I would say there's always a fun one if you like, in the AWS world, like, there's this thing called CommSum, and like, AWS Community Summit, and it's normally ran in Manchester, and it's ran by a really good consultancy, and and some other people. And the other people that run it, they run it because they really enjoy running it. Mhmm. Everyone runs it because they really enjoy running it. It's an amazing event, and they keep the ticket price quite cheap.

Mhmm. But it's not, like, it's a conference, and so you're providing value because you have really good speakers that you know, who you get to come to your conference, and you charge for people to come. Mhmm. And you make people pay to be sponsors.

Mhmm.

And then they get a little booth or whatever. That's that's one way of doing it, but that's a conference. That's not like an evening event. Yeah. And the conferences are where people can, like, make cash and, like, you can kind of do everything, but they're a huge amount of work.

Mhmm. Like, I'm I have a day job.

Yeah. This is

this is like a couple of hours a month we spend on this.

Yeah.

It's really really nothing.

Mhmm. So the the takeaway I'm hearing is like, because I think a lot of people think about running their own events, series as DevTools is Probably it's better to just find, well, you know, talk to you or talk to other Matt Carries and just support their event and just be a part of it.

Like Louie and Luke are tinkerers. Right?

Yeah. Like,

now they they were both DevTools startups. Mhmm. So well, not start not so much startups, but Louis is running Bloop, which is he's DevTool founder, and Luke is working at Growth at Eleven Labs. Both of those people running AI Tinkerers, they you don't see, like, Eleven Labs branding

Yeah.

Very much, like, around the place. Yep. Like, it's not an Eleven Labs event. If it was an Eleven Labs event, it would be a different demographic of people going. Mhmm.

And that's what's the most important, is like, you have to and you only really get one chance at this. You have to get somewhere, and the vibe has to be right. Like, the people who are there have to feel like they are augmenting you. Like, you're learning something from them, even just the people around you. And if you turn up and they're, like, a different style of like, the people who are there maybe aren't, like, related to what you're interested in, then it you're not you're probably not gonna get as much out of the event, and you're not gonna come back.

Yeah. 100%. And so that is actually the hardest thing about running an event, is knowing and I think this is what Luma's really helped with, the actual UX of Luma, is being able to be very selective about who comes. Yeah. I literally scroll through, and we we do admit people first come, first served, but we also reject people first come, first served.

Yeah. And that is an important distinction that maybe we don't shout about too much. Yeah. Can't we can't let everyone come. We just don't we won't have the space.

And even if we did have the space, we want the people who to come who are gonna get the most out of the event. The same way that everyone can't demo at every event. Like, I must get for the Stockholm event on on Wednesday. I mean, I've done, like, 10 interviews with startups about them demoing. There must I must have maybe 20 more emails.

It's like, it is a little bit ridiculous for for one potential spot that we had available. So, like, 30 to one is not great odds. And so it's your job as an event organizer to like craft that in the best way possible.

Yeah. You gotta

curate. You got to curate.

Yeah. Matt, I think we're probably at time. In fact, I think we're at time. Thank you so much.

Oh, dude. Thank you so much for having me on.

It was really fun. I learned a lot. I don't know if you wanna do any shout outs quickly before

Briefly shout. So the event series that I run is called AI demo days. You can find it at demodays.ai. And, yeah, I'm like Matt zed Carey. Carey spelled like Mariah at every on, like, every social media and GitHub and LinkedIn and Twitter and all that stuff.

Amazing. Well, thank you very much, Matt, and thanks everyone for listening.