Jose Saez-Merino spent more than 20 years in 3D and motion design before he wrote a line of app code. The reason he started was an Apple Watch Ultra he couldn't buy, and the reason he's now full-time on Matte, a Mac app for making videos of your app, is a timer he finished and never released. On this episode of Launched, Charlie Chapman follows the threads between them.
The timer app was done. The video was the problem.
After iOS 26 and Liquid Glass, Jose started playing with shaders that made liquid bubbles and metal balls, and ended up with a timer that drips liquid out of the Dynamic Island. Charlie remembers the tweets. "I have that app is done. It's ready," Jose says, and Charlie fills in the rest: "You just never released it."
The reason had nothing to do with the app. "I can't make the videos I want to make about this app because every time I need to record a demo of this app or make a video. I end up having to use four different apps." QuickTime or the simulator to record it, Final Cut or Premiere to cut it, After Effects for the effects, and Cinema 4D if he wanted anything in 3D. "It was a long, boring process of making a cool video about your app. So I'm like, well, the obvious solution is I'm going to make my own app for making videos, right? Like this is the obvious solution."
He'd tried a version of this once before: a website that pre-rendered 3D phone videos and corner-pinned your screenshots into them. It worked, and he shelved it the day the work turned into making templates. "Building the system is fun," Charlie says. "And then filling out the system is grunt work, right?" The difference the second time was that Jose needed the tool himself, every week. "If it's something I wouldn't use myself, I'm not going to put the effort that I need to put in to make something good."
It shipped in January with flat frames, no beta, and paying subscribers
Matte's first version was, in Jose's words, "Super basic." No beta. "I just launched it. First version." The phone frames weren't 3D; they weren't even Apple's official frames, "just like a vector version of the iPhone frames." What it had was the part the screen-recording tools were missing: you could record from the simulator or a device and then cut, reorder and trim the clips without leaving the app. "To me that was the important part."
The rule he took from it: "it's better to ship fast, even if it's early, even if it's not what you want it to be, and then start building on top, because you're going to start getting some clients and some customers and people are going to learn about it." By the end of January, "People were subscribing." He'd also just started a new day job at a startup lab the same month, which tells you how rough the first version was allowed to be.
Claude opened Blender and rigged the 3D hand itself
The feature people kept asking for was a hand holding the phone. Jose had been putting it off for a while, because he knew the work involved: find a hand, pose it, work out how it fits in the product. Then he bought a rigged hand model with decent textures and wrote one long prompt. "The prompt was like a big paragraph, I want to do this. This is more or less how it's going to work. We're gonna start prototyping it, and I want you to build a first post, which is basically just the hand holding the phone. And it just went nuts. It went nuts."
Claude opened the hand model in Blender, measured the sizes of every phone Matte supports, and wrote rules so the grip would work on all of them, adjusting the bones of the hand for each size. "I think it's like less than an hour." Jose opened Matte, switched phones, and watched the hand adapt. "It blew my mind that I was just able to do that in basically one prompt." The iteration that followed was art direction rather than modelling: "these two fingers move them a bit downwards and then twist the knuckle a little bit more. And it would just it would just do it."
Jose's name for what's happening to his job: "you're turning from an artist to a creative director." He's not sentimental about it. After 20 years in video and motion design he says he can see every flaw in what the models produce, "but I can also see how it's getting better release after release after release." Charlie's take on where that leaves the rest of us is the sharpest line in the episode: if two people with the same model build the same app and it looks the same, neither succeeds. "It's your preferences, it's your creative direction. It's your thousand no's for every yes."
A 20-year video editor had to unlearn video editing
Matte's users are developers, and "most developers aren't familiar with video editing software in the same way I am. So one of the things I've had to do is sort of like unlearn everything I know about video editing and motion design for the last 20 years." The concrete case is scrubbing. In Premiere you click the playhead and drag it. In the newer tools, "you'll just hover the mouse over the timeline." Jose left hover scrubbing out, "and one of the first feedback pieces I got is why can't I do this?"
Charlie, who has a video editor of his own in progress, had turned hover scrubbing off in his editing software because "it drove me nuts." Jose's answer is the one most indie developers land on eventually: adapt, then "make it an option." Matte now defaults to hover scrubbing with a switch in settings for people who want the old timeline. The same instinct shapes the rest of the app. Drop in a screen recording and "you need to give them something that could be shareable immediately," because most people using it "just want to go in and build their video. Get out."
The marketing strategy is the one he's sorry to report
Jose gets asked about Matte's marketing often enough to have a stock answer. "I'm sorry to disappoint, but the only thing I do is talk about the cool stuff I'm making, because that's the way I work. I get excited about whatever I'm making. I post it on Twitter, and that usually works a lot as marketing." It's been the playbook for every app: his first, a stock monitor for iPhones, launched through Twitter and the Spanish tech press he knew from his years as a phone-review YouTuber. His second, a widget that says yes or no to "did I lock the door," got teaser images and videos on Twitter, then a post on Threads where someone tagged MKBHD. "He replied and that like it made it explode." It had subscribers immediately, and Jose sold it about a month before the recording because he had too many other things going on. "I honestly, I was bored of it."
Matte got the same treatment and took off from the early videos. "That's my playbook. I just talk about my stuff." A week or so before recording, Jose announced that Matte is now his full-time job: the app is profitable, and "it's getting better every month, so revenue is increasing every month." It's never run an ad.
Also in the episode: why Jose refuses to call a web app an app, the 1-a-month-or-25-forever pricing on his widget app and why Northern Europeans bought the lifetime, the MCP server bundled inside Matte so agents can drive the editor, and the model-subscription shuffle that leaves both of them with 2% of a quota.

