Tyk’s founder and CEO, Martin, wanted to reward technical innovation by introducing the Tyk Side Project Fund. After all, he knows what it’s like. He started Tyk as a side project when he couldn’t find the ideal API gateway. He had to make one himself, experiencing first-hand the highs and lows of taking a side project to a global enterprise.
We are thrilled to announce that Tabby, an AI browser extension which helps you declutter your tabs, is one of this year’s winners. Here’s François Meunier, one of the three founders, to tell us more…
Tyk: Tell me about your side project. How did you come up with the idea?
François: Basically, it’s a browser extension that solves the problem of tab cluttering. This is when you have 200 tabs open in your browser and you have no idea where the one you want is and you spend a lot of time going through them trying to find it.
The idea came from my work. I would be in an online meeting and my manager would share a screen and want to show me some slides. There were so many tabs open that you couldn’t even read them, they were so small. At first I thought that maybe my manager had a problem that I don’t have – he doesn’t know how to manage his tabs. But actually a few weeks later I started to realise that I have the same problem! I talked about it with my friends and we decided to try and build a solution because it felt like we were all struggling with this. We don’t think people are necessarily aware of it, which is also an issue, so we just tried to tackle that problem.
This is what led us to create Tabby. Tabby uses AI to try to understand which tabs you’re using and which ones are not useful anymore. When we think that one of your tabs is not useful anymore, we will close it, saving it for you so that you can open it again if you want.
Tyk: That sounds like such a great idea!
François: Yes, I think we are tackling it in the hardest way though! When we pitched the idea, some people said, “I don’t want my tabs closing, I know why I’ve got these tabs open!” So it’s a bit tricky to convince people to try it, but it’s looking promising. My brother is a designer and I’ve talked to him about it. He was telling me that in his line of work he keeps a lot of tabs open because he knows he needs to look at them for inspiration. So, in the end, it’s like using the tabs as a bookmark which is not the initial intent.
Tyk: How have you tested and validated Tabby?
François: We were always conscious about privacy, so one of the big things about Tabby is that we are not going to retrieve anything that’s going on in your browser. However, this can be a problem because it means we don’t have visibility on what’s going on if something goes wrong for a specific user. To solve that, we have a pool of testers who help us monitor if things aren’t going right.
Another thing that we plan to do is create another version of Tabby. This would be a private version just for us, which we would use with testers who agree to us viewing their data. We can then analyse their data, which would be a more robust approach.
Tyk: What have the main challenges been?
François: Feedback is the main challenge for us. We have an email feedback form but even with that sometimes we don’t understand why it’s not working because the information is not specific. We are using statistics but we don’t want to force people to share data with us so we are looking at how to do it better. It’s not simple but we really want to be transparent with people.
We also realise from talking to people that if we start saying that we’ll know everything they’re doing online, that would be a huge problem! So we need to find the correct way to anonymise data and to present the opt in or opt out options.
Tyk: What stage is the project at currently and what are your next steps?
François: The project is growing by itself right now. We put a lot of effort in over the last year, trying to make it grow. We have around 2,500 users, so it’s growing slowly but surely. Now we’re waiting for more users, so we can start putting more work into it because there’s a diminishing return for us. We have put a lot of money into it so far, and now we need to see whether users are enjoying it and finding it valuable. The best way to do that is to wait and see. I would say if we reach 10,000 users then we would commit to developing it.
Tyk: What difference has the Tyk Side Project funding made?
François: Two weeks ago our Gmail account stopped working. There are so many things like that which need monitoring. There are three of us working on the project, but we all have busy, fulltime jobs so finding the time to sort out our website, for example, is challenging. The funding gives us peace of mind that we can get any administrative problems sorted out.
Tyk: What are the values that drive you?
François: I’m a really technical guy, from my studies but also I’ve always been into technology and the internet. I think that we need to make the internet as simple as possible for people and, generally, any technology should be really simple to use. Companies like Apple did an amazing job doing this but I think there’s still so much we can do.
The three of us at Tabby are deeply into machine learning and AI. I don’t want to fall into the idea that AI is going to be crazy and kill all humans but we think that we have to find a way for AI and humans to work together. AI can and should act as an assistant to humans.
With Tabby, what we want to say to people is, “Hey, you don’t have to close your tabs yourself. Your computer can do it for you.” That’s the main thing about Tabby – you don’t need to do that boring thing yourself! Your computer can learn it and do it for you.
Tyk: What tips would you give to someone at the start of their own side project?
François: I think the most important tip is to find really good testers. That’s the thing that helped us validate what we are doing. It was really hard to kickstart the project in the beginning as it was a pain to find users to install Tabby! So, yes, find lots of people who are willing to try it, even just for a week – let them try your project for free and give you feedback. That’s really key, because you have no idea if what you’re doing is good because you are so biased about your own project. You should always try and get feedback, even from people you don’t know. We did this a lot and it was really valuable.
Tyk: Where will Tabby be in five years’ time?
François: One direction would be to see Tabby as an official extension of Google Chrome where you could set your tabs by colour, for example. Extensions like Tabby are initially bought by enthusiasts but to see it officially endorsed by Google would be amazing!
And the other direction is to see how it goes and if people like it. We are also considering a plan to go totally open source so that Tabby would become a community project.
Tyk: Is there anything else you’d like to add?
François: I’d like to thank my two partners and everyone else who’s helped me with this project. It’s been a lot of fun and I’m really looking forward to seeing where it can go.