Pixel MBA

PixelMBA is a student from Portugal who needed a creative way to fund his studies. Rather than run a standard crowdfunding campaign, he had a different idea: sell pixels. Supporters would buy space on a canvas, customize each pixel with a GIF or image, and leave a personal mark while contributing to his education.

The concept was compelling but technically non-trivial. It required a real-time purchasable pixel grid, per-pixel customization, secure payments, and user accounts, built from scratch and delivered as a production website.

Cenciss took the project from requirements to live deployment.

The pixel donation system was the central engineering challenge. The canvas needed to track individual pixel ownership across all users, prevent double-purchases, and display customizations accurately in real time.

The customization feature added another layer of complexity. Each purchased pixel had to accept a unique GIF or image from its owner and render it correctly at small scale without breaking the overall grid layout.

Payment integrity was non-negotiable. Purchases needed to be atomic: payment confirmation had to trigger pixel assignment, with no risk of a paid pixel being left unassigned or assigned without payment.

User authentication was required so donors could return, manage their pixels, and update customizations after the initial purchase.

The platform was built on Node.js and MongoDB, with EJS handling server-side templating and rendering. MongoDB stored pixel ownership records, customization data, and user account information, with the data model designed to make per-pixel lookups fast and consistent.

Stripe was integrated as the payment gateway. The purchase flow was structured so that Stripe confirmed payment before pixel ownership was recorded, ensuring no pixel could be claimed without a completed transaction.

User authentication was implemented with session management, giving donors a personal dashboard to view their purchased pixels, update GIFs or images, and track their contribution to the canvas.

The customization system accepted image and GIF uploads, processed and stored them against the pixel record, and rendered them dynamically on the public-facing canvas.