I built a full 3D smart city monitoring system… with just AI prompts. No traditional coding.I built a full 3D smart city monitoring system… with just AI prompts. No traditional coding.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
I built a full 3D smart city monitoring system… with just AI prompts.
No traditional coding. No complex pipelines.
Introducing — Cortex City
A real-time urban intelligence dashboard running entirely in the browser.
🌆 Generated an entire city from scratch — bridges, skylines, parks
⚡ Monitor live energy grids + detect brownout risks
🚦 Track traffic density with dynamic flow visualization
🌧️ Analyze weather patterns and storm activity in real-time
🚨 Emergency Mode — city-wide alert system with scan waves + radar targeting
What if you could explore a basketball arena in 3D before booking your seat?
You found the game. But before you pay, how do you know which seat is worth it? A flat seating chart gives you a section and a number, but it can be hard to picture your position relative to the court.
So I built BOUNCE with Lovable, a basketball ticketing concept centered around interactive 3D seat selection.
Rotate the stadium, zoom into the stands, explore different sections, and compare prices before choosing your seats directly inside the 3D arena. The goal is simple: help fans feel more confident about where they’re sitting before they book.
From there, complete checkout and get a ticket you can view, save, or print. My Tickets brings bookings together, with a one-time seat change for the same game and refund requests for eligible bookings.
Bold orange, oversized typography, and scroll driven motion bring the energy of basketball into the experience.
This summer my dad wanted a new fence and deck. The quotes took over a week. Looking at it from the carpenter's side, the hard part isn't the calendar, it's driving out to measure before you can even write a quote.
So I built Karlsson Staket & Altan, a made-up solo carpenter in Kristianstad, Sweden, where the customer does the measuring by drawing.
For the customer:
Search the address and draw the fence or deck on a satellite map – length and area are measured automatically
Shape it in 3D: height, material, railings, stairs
Instant price with the Swedish ROT tax deduction
Pick a build week with live availability and get a booking reference and a calendar file
For the owner
A 12-week season board with drag-and-drop and capacity per week – full weeks disappear for customers
Material lists calculated from the drawing, merged into one order with stock deducted and rounded to full packs, plus a shopping list
One page per project: quote, booking confirmation, work order and draft invoice
Adjust the customer's 3D model and send it back with a comment to their personal project page
How I built it
I planned first: chat mode in Lovable, a starter prompt written with Claude Opus 5.5, a hand-drawn design sketch, and a plan I approved before any code. Version 1 stopped at a price, nobody actually got booked, so I kept iterating in small steps. The hardest part was the 3D model. It was bout 10–15 hours in Lovable, using Lovable Cloud and Google Maps, with prices recalculated on the server.
Try the guided tour: https://karlsson.lovable.app/demo (demo mode, nothing is saved)
Note: the project photos are AI-genera::ted, and the before/after numbers on the site are estimates, not measured data.