Brickroad

01 / Brief

Solution

Brickroad is a data marketplace dApp where dataset owners can upload and monetize their datasets, and AI companies can buy the data they need. Before this kind of workflow, many users relied on platforms like Kaggle and did not have strong ownership/control over monetization of their own data. I worked in a 2-developer team, with full ownership of the backend and partial frontend integration work. I built the backend end-to-end: dataset segmentation, dataset storage on Irys, data visualization services, and payment flows with Stripe, while integrating wallet management with Privy. The goal was to deliver a full MVP pipeline where users could deposit datasets easily, store them on Irys, buy dataset portions, and get results quickly.

02 / Constraints

Problem

• Keeping dataset segmentation performant on large uploads
• Ensuring decentralized storage reliability on Irys
• Maintaining fast and stable upload performance

03 / System

Architecture / workflow

• End-to-end backend delivered in a 2-developer MVP project
• Dataset uploads and decentralized storage on Irys
• Dataset segment purchasing powered by DuckDB
• Payments integrated with Stripe and MoonPay
• 100+ dataset uploads

04 / Evidence

Results / impact

• Shipped a full dataset monetization MVP in 3 months
• Delivered backend services for segmentation, storage, visualization, and payments
• Reached 100+ dataset uploads

05 / Gallery

Visual references