Sabba Keynejad, CEO & Co-founder at VEED.IO

0:00All right, we are back. Thank you for joining us, Sabba. It is absolutely amazing to have you. There's very exciting stuff going on with Veed at the moment. We are very excited to get into all of it. But before we do, why don't you just give us a super quick intro into who you are and what you're building? Sure thing. Well, thank you so much for having me. I really appreciate it. So, yeah, I'm Sabba. I'm CEO and co-founder of Veed. Veed is a video creation platform. We're headquartered here in London. We've got about 10 million monthly users and, you know, four or five million videos created every month through the platform.

0:30amazing you're absolutely killing it 10 million users yeah insane and who are those users so if

0:37Sabba Keynejad:i don't know vid which i do because i love uh using using video tools but imagine i don't who should be using you or why should they be coming to vid sure so i mean um vid really focuses on making really great talking head videos um anyone that's got a message wants to share it and you know video is definitely the best way to do it right now so most people use us for marketing and social media they want to make that talking head video make it super engaging great subtitles or potentially they want to generate a video to do a very similar thing as well so yeah if you're in marketing want to get a message out there uh that's definitely video is the place to go and the other thing that we focus on a lot is like simplicity and like ease of use so i think so many video creation platforms even getting the av set up for this for example can be a little bit complex and we want to kind of reduce that barrier and allow more people to be able to do it.

1:28Sabba Keynejad:And so it's both on the editing side and on the Gen AI side for talking heads. Exactly. I mean, you know, recording yourself can be a bit daunting and sometimes having AI to do that for you is actually quite useful. It also allows us to allow anyone to create a video who's got an internet connection access to a laptop. They don't need to be on camera and I think that's quite important for you know making it more and more accessible super cool and um what what's like kind of the hottest feature right now what are people wanting to do in the space of talking heads right now the most honestly it's really hard to answer that question often when you're speaking to your users and you're trying to work out what's the next feature what should we invest in what should we build they don't know what they don't know so often the things that they ask for is a better version of what you've got which isn't quite that it's not super useful so often what we have to also do is spend half our time doing r &d for example um and and kind of thinking hey what's next what could really improve these flows something we launched literally uh on monday was a video generation model for making talking heads that went crazy like the launch we've never seen anything like it um you know the videos that went out on social for example was like two million views in 24 hours.

2:42So yeah, that's a good example of, hey, people don't know what they want always, so we have to do a little bit of experimentation. But yeah, that's how we see it.

2:52Sabba Keynejad:When they see it, they know. When they see it, they know that they wanted it. You know, from the notes and what we're going to talk about today, this is definitely later in what you sent me. But I think like, you know, there's two ways to look at it, right? There's like capturing demand which is more about kind of hey someone's got a problem or a job to be done how can we capture that demand that people we have but then there's also generating demand showing something cool that they might not even know they want or but but you know something that can fit into a workflow they're doing you have to kind of invent it for them and show them hey we can do this and this this could help your workflow and has that changed over time like have you focused in the initial stages on like uh servicing that demand and now you're trying to push the frontier out or have you always trying to be like okay let's build something that people don't know what they want yet um i used to have this saying so i think yc have this thing called make something people want it's like their famous mantra right yeah and i kind of adjusted it when we started the company to make something people search for and i was very obsessed with like capturing demands that kind of was already existing.

3:57And it was really powerful, actually, because it also informed our roadmap. Like, we'd kind of build the next feature based on how much search traffic there was for it, because that kind of indicated the demand that there was for it. So that's kind of how we started. It was very kind of scientific in a way. It was like a process. And I think, you know, when you're number one for every single, not every single search term, but like if you Google AI video, we should come up number one. Where do you go next? That's when the generating demand comes in. And I think that's kind of where we are as a company now and kind of how we're thinking about marketing as well.

4:30Very cool. And is this, you know, what you're talking to, what you're kind of alluding to, is that Fabric 1.0? Yes, exactly. Yeah. Can you touch a bit more about like what that is and where it's come from? Sure. So, you know, as I said earlier, most people use V for making talking head videos, but always being on camera or hiring influencers to make videos for you can be a little bit, you know, time consuming. I mean, if we also look at like the most popular videos that people are creating with things like VO3, they're just talking videos. Like people love talking videos. So we thought, hey, wouldn't it be cool if we could make a model that specialised in doing exactly that?

5:05It was designed specifically for talking videos. And the way that we kind of thought about the speed of the model, the quality and how much it costs for us to generate. We were very like hyper focused on building it to be very performant and very cost effective so we can roll it out to our user base. so fabric one is basically a diffusion uh based video model that kind of takes an audio um stream and then lip syncs that with the emotion to to the image that you give it for example if i was to have a audio of myself whispering the model would actually make me lean in and whisper if i was shouting it would make me very animated um so it's incredibly powerful model and uh yeah it's you know we think it's going to be really important for how people create videos and i'll give you a really, really interesting like use case.

5:52Like I was on holiday last week, the team wanted to make a video for the launch. I literally just sent them an audio stream, an audio on like WhatsApp. And they made a really high quality video of me for the launch and I was never even there. And the quality is incredible. So yeah, super cool.

6:07Sabba Keynejad:What is needed in terms of input to make that happen? Obviously of you, they'll have a lot of footage. What does the average user need to do that? Well, it depends if you wanna have like a fine tuned version of yourself or you just want to use a stock character so the easiest way to do it is go to the site click a pre-created character that we have and then that's mapped to a voice uh if you want to do one of yourself it's just an image of yourself and an audio oh wow that's awesome very critical yeah easy than being on camera super interesting yeah it's interesting i think like i was chatting with a friend about this earlier like I think we're moving to a world where AI is a new camera.

6:47And people aren't going to be super... I don't know. Right now, it feels like there's AI content and there's non-AI content. They're two different things. I think that world is going to merge. I think people are not going to be as... You know, like scrutinized that much.

7:03Sabba Keynejad:Yeah, we were just talking to a previous guest about this. How, I mean, Seb and I both love to create content. how today we're still in a weird spot where sometimes just pointing the camera at yourself is easier than going the AI route if you're not so familiar with the tools, because we're so used to that. And so that's where there's a divide. But how do you think those two sides will kind of grow together exactly? Well, I'm the same as you. I've got no problems whacking my camera out and just rambling on. And I don't care that much about how I look on camera. I just get it done. um but so many people just don't like i think more people don't than do right um so yeah i think like i think you know with in the next 12 months we're probably just going to see a world where like you can fine tune yourself very easily with a model and then put yourself in different environments backgrounds based on what you need to what you need message you need to share and often with like talking heads and AI it's often seen within like the business use case in LND but I think we're going to start seeing it much much more in like marketing and on people's websites and on social media as well yeah I mean I'm I've recently started doing TikTok you know and I've always done content creation on LinkedIn which is always like written that's always been the thing that I like doing but now what I'm doing is I'm taking my top performing LinkedIn posts and I'm recording a video of myself and slapping it on TikTok with like a green screen or whatever but for me I feel like perfect use case where if I could just give it the script of my LinkedIn posts I wouldn't even have to film myself right I could just you know use Veed use my own talking head and then just have it created for me basically yeah I think I think the we should also like not think that this is like a you know you can go to chat GPT and say hey write me a blog post about you know scaling Europe for example and it would do an average job to it and no one's going to want to read it it's similar video you could just churn out video after video after video but like the i think creation right and like editing really starts becoming the craft to make things important that's that's always going to be the case always going to be the case i think tasting creation becomes incredibly important very important if you can just create anything then yeah it's really picking and choosing what what you should create um yeah so it's easier right it's easier but then that storytelling aspect or that creation aspect is the thing that really, really makes your content different.

9:34And how are you thinking about content creation and storytelling in, you know, I guess your own journey, your own growth? Has that been something that you've tried to leverage a lot of to kind of get you to that 10 million users? Yeah, I mean, we spend, which was quite weird, a long time, and we still do, making YouTube videos. It's been a really big part of our growth journey we i mean google's the sorry youtube is the second largest search engine in the world and often there's there's some things when you want to get an answer to it you want to kind of just read that answer but a lot of the times if it's using software you want to see it being done in front of you so you can mimic it right so we've been plugging away at youtube for years and have amassed about 300 000 subscribers across a couple of channels and those you know channels do over a million views a month and it's been really important because when you're kind of you know say for example someone wants to trim a video or crop a video and they want a tutorial hopefully a vid video comes up in that journey but you're not just educating them how to do it you're also activating the user at the same time so it's a very powerful mechanic and yeah trying to eat our own dog food there as well and you know use our own tools okay speaking of like growth

10:40Sabba Keynejad:hacks and getting to where you are now um you bootstrapped at the beginning correct correct We would love to learn a little bit more about that because, you know, a lot of there's a lot of capital in the market, a lot of founders raised before having anything. That's almost controversial to start with Bootstrap. We would love to learn why you made that decision and what you think that taught you. So I first want to be very clear. Originally, it wasn't a decision that I had. It was, you know, it was like there was no other option. you know when we started the company i think we were just very scrappy and when we tried to raise funding it wasn't the right time we tried to get into yc that didn't work and we were just a bit like you know let's just not go that route basically let's let's just go to the route where we're making like 10 20 000 a month and this is our and this becomes our jobs right that was kind of the goal but the company grew incredibly quickly it was like i mean for for then Like now a company growing quickly is like 100 million ARR in six months, right?

11:44But like back then it was like 1 million in the first year, got to 6 million in the next year and it was growing incredibly quickly. So we didn't really need capital at that point. When you start, you know, generating revenue, being able to attract users at a really, you know, fast and efficient rate, it just really works. So a lot, it's interesting though, because a lot of what we learned then is kind of still in our DNA. we're still very capital efficient we're pretty much a profitable company like we don't burn any money so it's it's that that dna has kind of always stayed within us and is that affecting how you're looking at the future are you thinking about raising again down the line or are you happy being kind of at that sort of like profitable level i'm very happy being there yeah it's um it's it's good to control your own destiny right and i think that yeah i think i also i also think that there's a lot of overhead when you have such a large team communication overhead innovation overhead and it's just nice to be able to you know be be be in founder mode so to speak and get things done and like be close to the engineers and be close to the product and uh i've really enjoyed it that said never say never you know you don't know what's around the corner yeah very cool and when you did raise you raised from sequoia right that's like uh and they must have been so satisfied after like yeah so we raised from sequoia probably three years ago um and it was a 35 million series a it's still the first ever round that we've we've done um but it's been a great experience and we've really loved working with the team there and i think arguably we've built a much better company because of it as well we've got way more disciplined as well i i think and you know and it's also great for attracting talent and people's believe in you you're kind of like able to have that brand halo sequoia rub off on you and people in tech know what you're talking about when you say you know that they're your investors so it's been it's been really good and it's also i think one of the first investments out of out of the london office as well which is really cool yeah somebody told me that which is yeah yeah which i think is like an amazing success and classic sequoia to like sort of like hit it out the park on their first one uh but what was that process like raising from them were you able to go with them and you know show this like bootstrap journey did that story really really appeal to them that sort of like scrappy scaling fast but like you know making it work without the funding was that what kind of got them on board like how were you able to convince them?

14:09Yeah I can't say from from you know their perspective but I think from what I've learned is you know if you if you do the work and your numbers are good a lot of people will be very interested. I was uh you know because we were kind of bootstrapped and maybe I felt a little bit, what's the word, burnt from the experience. I was just posting our revenue on Twitter like every other day. And so it got a lot of attention. And when you can see the growth rates that you're doing month over month, a lot of people start knocking at your door. So I think a lot of people knocked at the door. So, and also I think there's a part of, you know, I was like, I ain't doing this, I ain't raising.

14:49And I think that that also works as well so i think um yeah just grow show you've got some good numbers and i think a lot of things fall into place you know yeah very cool it's a great story uh i wanted to ask about your your your recent hire of i think pete steinauer yes i've seen that right he's like an uh an apple vet i think he's been there since 1992 in the video application space what was the thinking like you know why have you chosen to bring someone like that on a cto and how did you convince him to leave a company that he's been at for 30 years to come and join your mission? You know, sometimes the stars just align.

15:28You know, I'm a believer in, you know, the world just kind of aligning for you sometimes. And I think that was definitely one of those cases. I think, you know, building Veed is definitely more on the, it's technically a really challenging product. And there's a lot of technical expertise in the depths of videos when you start looking at codecs and transcoding and all this good stuff. And how do you structure an org to build these products there's not loads and loads of information out there about how to do it so we really wanted to find someone who just knew what they were doing and who'd done it before and could like help and advise the company also if i look at the dna of the company we're definitely like the scrappy startup and that's great but it's not always great sometimes you need a very stable product right and you need someone who knows how to architect it in a way and so we're trying to find people that had experience basically doing this and obviously we got led to uh apple uh because you know they've got some of the most popular video creation products in the world um and yeah as i said stars stars really did align and uh it's yeah it's been a great journey uh working with pete since he's since he's joined amazing yeah yeah i was looking at his linkedin he's been i think in in apple for 30 years but in video for 15 years um which is yeah there's probably there aren't that many people who have been working in the early days of video Yeah, I think there's only one person actually, and it's him.

16:49Yeah, yeah. It's like an amazing hire. I remember I spoke about it on LinkedIn because we are seeing this trend of, you know, big American tech talent actually coming this way to Europe and actually joining some of the young startups that we're seeing, which is a really interesting trend. But very cool. It's going to be super exciting to see kind of what happens. There's a lot of cool things happening in Europe, right? And there's a lot of great advantages for being here. So, you know, I think we're going to see more of it.

17:14Sabba Keynejad:for sure absolutely maybe maybe one last question from my side what is the next iteration what's the next evolution stage of Eid gonna be yeah I mean I don't want to sound like every other person that's going to say the word AI but unfortunately I have to say that just for one main reason every single green shoot that we have on every single graph is rooted in AI and generation um i think we're going to be pushing a lot more into creating our own models but also allowing people to interact with other peoples as well and just really trying to push the boundaries of what you can do with video generation the demand that we're seeing and the and the pull from the market is really really interesting uh and i think we're also going to start experimenting very soon i think next week we're launching a one of our fabric model via an api as well so let's see what happens to that side of the business but i think it's really exciting and i've never you know since the launch there's like 100 people in my dm saying can i have an api so having that kind of like pull yeah from the market is normally a good sign and uh yeah let's see where we go who who knows what's around the corner but look it's it's it's an exciting time to be building right now yeah definitely very excited to see what's what's next no this is amazing but let's say i'm conscious we're just over time um but like things are obviously going really well for the whole team and yourself over at video it's like super exciting time to be building in this space we're going to be looking at the releases, using the tools ourself.

18:41I'm going to be like getting on it as soon as I can because it sounds amazing. But thank you so much for joining us. And yeah, we look forward to kind of seeing you grow even more. Amazing. Hey, I appreciate you having me. Thank you so much. Thank you.

18:53Sabba Keynejad:So great to meet you. Have a good one. Bye-bye.