Hey all, I'm Nejat, an experienced product designer from Turkey.
I've been sitting on this idea of a Jiu-Jitsu focused map where you can browse gyms in Istanbul and get more information compared to Google maps, only to lower the entry barrier for the sport a bit more.
Whenever I thought about it or brainstormed with my co-founders (a small BJJ gym here in Istanbul) it usually ended with "we'd need a developer for that though".
I studied software engineering and worked as a jr dev for a year so I have technical background but I can't really code any project from scratch. The only other community project I built is BJJ Resources, which is a simple directory site built in Framer. I have decent experience with no code tools but this one was a bit more complex to be built with such tools.
About a month ago, I just wanted to see what this "vibe coding" is about and how capable the existing tools/models are. I wasn't really expecting much, I thought it would require a lot of configuration with map service APIs etc. Also, I always thought even though it's AI assisted, I would still need to write the code. After trying lovable, bolt and v0 with the same basic entry prompt, I decided to stick with v0 since that produced the best looking UI as a baseline.
I honestly didn't really have any solid plan for this project. I normally would sit down, brainstorm and explore the design side of it but since it was supposed to be a test, I dove head first without expecting much. Aside from tweaking some values in the code here and there, I didn't have to write any code.
Fast forward +300 iterations, I published the website last week so I figured I might post in this community as well: https://www.istanbulbjjmap.com/en
The most challenging part of this project (other than getting accurate info from gyms) was the URL routing. At first, v0 just built it with simple modals and I thought it would be fine. As it came closer to the finish line, I realized I will need proper URL paths for the gyms, both for SEO and better analytics. One thing that I knew was I didn't want gyms to have their own pages, I still needed the profile to open in a modal when the link is visited. I was naive to expect that it would be an easy switch but I was very very wrong. Got help from Claude, broke a lot of things, fixed many bugs but it's working properly in two languages now (TR and EN).
Vibe coding genuinely brought back my entrepreneur spirit back. The best that I could build on my own was marketing websites with fancy animations but now I feel like I can build pretty much any idea of my own. Maybe not the next facebook or airbnb but I never had such ambitions anyway.
It's truly an eye opening experience as a designer to explore possibilities with real, working prototypes especially when you do it by just describing the steps. Lucky for me, this project doesn't contain any sensitive data so I didn't really have concerns regarding security (maybe I should be? idk, let me know if you have any warnings for me)
I'm already working on 2 other projects, one of them being an internal CRM. Definitely gonna build many more!
