[Tech-a-Tete] “We didn’t want to create another Instagram” – Alexey Moiseenkov, CEO, Prisma Labs
“When we launched Prisma, we were just five people”
He is the CEO of the company that gave us Prisma, and more recently, the Sticky AI app. But what exactly set them on this path? How did they get here? And where are they headed? Alexey Moiseenkov shares all this and more in an exclusive interview with Technology Personalized.
You have just launched a new app called Sticky AI, which turns selfies into chat stickers. It seems a bit of a change of track from Prisma, which converted photographs into works of art. Why this change?
We experiment a lot! Sticky AI is a very simple app for creating stickers. We believe that AI can help people express themselves more precisely. Our goal is to help people turn their feeling and emotions into digital creations very easily and quickly. We live in the era of messengers, that’s why we want to give people a way to react on and give feedback for every message they get. That’s how AI is going to make the way we share our emotions better. Sticky has a great potential and we’ll do our best to turn it into a cool app for self-expression.
There is a perception that with Sticky AI, you are trying to get on to the Snapchat/ Facebook sticker bandwagon…
Absolutely, not. Our technology that allows us to detect a person and cut his background is unique and works completely offline! We are also able to detect any other smaller objects on a photo and video (even on a live video stream) and work with it. So Sticky has a very complex tech behind a very clean user interface. There are very few teams in the world that are able to show great results using this technology.
Let’s go back to the app that made you famous and a question that you must be getting asked very often indeed: what inspired you to make Prisma?
We wanted to bring an ultimately new way of self-expression and include a very complex tech there. We really wanted people to have the most advanced technology in their pockets. Style transfer was very promising and similar products were really slow. People had to wait for 20 minutes to get a picture processed. We managed to fasten the process (thanks to our algorithm) and that was a great breakthrough.
How difficult was it to make Prisma? How long did it take and how many people were involved in the effort?
When we launched Prisma, we were only five people. We fully developed it within a couple of months. The most difficult part was to apply some math tricks to the original algorithm to make is very fast.
What do you think made Prisma the rage it became? Were you surprised by its popularity?
[pullquote]We didn’t even spend a dollar on marketing or PR[/pullquote] We surely were surprised! We just put the app on the App Store to test our servers capacities and it went viral in a flash. People started sharing pictures on Instagram and telling each other about the app. That was very crucial. Nothing is better than people telling each other about the app. We didn’t even spend a dollar on marketing or PR. Also, every news media wrote about Prisma. That helped a lot!
Why did you release Prisma on iOS first? There are more Android users, surely…
We wanted to test the app at first. That’s it. We had a very small team and no Android developer at first. That’s the reason.
One thing that I have noticed about Prisma is the frequency with which you update the app. There was a demand for offline processing and you brought it, you keep adding new filters…and so on. How do you decide which changes to make and how difficult is it to make these changes?
We have a very big list of things to do. They include big and small updates. For example, we were the first team in the world to implement neural networks onto devices. We created our own framework and made Prisma completely offline. No one believed it was even possible. That was a big update and we had been working on that for a long time. Small updates include daily styles, fixes, improvements and other features. We decide what to do first depending on what users need. And we also experiment a lot. Offline also helped us save money on servers and make the user experience the same no matter where the users were and which type of connection they had.
We have seen Prisma evolve from being just an image editor to something closer to a community. What has been the reason behind this evolution?
We just wanted to have some place where people could keep their artworks. That’s it. We didn’t want to create another Instagram.[pullquote]We have 10 million active monthly users right now[/pullquote]
Some people think that Prisma after its initial surge of popularity has sort of lost momentum. Are there any statistics you would like to share with us in this regard?
Our user base has been pretty stable for the last six months. We have 10 million active monthly users right now.
India has a lot of app developers, all searching for the next killer app. What do you think makes an app successful? Any advice for them?
Something completely new with an easy user experience and simple design. It’s really hard to give advice here.
Right, simple question: what are your own favorite Prisma filters and why?
Tears, Mononoke. These are one of the first filters we had and they look great!
What’s next for Prisma? And Sticky UI?
As always, we’ll try to make the apps better implementing new features which I cannot talk about right now, sorry :)