Crafting a peer-to-peer marketplace with NFTs as collaterals




November 2022 - December 2023


Web Application





Zexus Finance needed a platform capable of directly connecting borrowers and lenders, enabling them to exchange ERC-20 tokens using NFTs as collaterals. This groundbreaking idea required a robust design and the leveraging of smart contracts to secure the borrowed funds while offering a smooth and efficient financial experience.


UX Design
Interaction Design
Design System
Product Discovery



Introduction and Goals

While the existing peer-to-peer financing platforms can speed up transactions when borrowers and lenders align perfectly, it usually requires extended communication and adjustment of the offer before a final agreement is reached. Zexus has partnered with 10Clouds to remove this bottleneck.

Recognizing the vast potential of NFTs and their diverse functionalities, we aimed to address the growing demand for NFT financing by offering a user-friendly, one-stop solution. We tackled the "matching speed" challenge by adopting a secure peer-to-peer model, profoundly streamlining the borrowing/lending process.

Our initial goals included having a platform MVP ready within two months. As a result, a robust marketplace that would connect borrowers and lenders seamlessly has come into being, equipped with functionalities that empower informed financial decision-making.

Additionally, we focused on establishing a strong brand identity to solidify Zexus’s position within the NFT financing landscape.


For our solution to work out, we had to merge a blockchain-powered back-end with a well-designed and reliable front-end, so deals on the platform could be done quickly and comfortably.

Developing the platform presented several major challenges. The first was adhering to a fixed budget, which required constant prioritization and tough decisions regarding which features were most crucial for the initial MVP. This careful balancing ensured the project stayed on track while delivering a solid foundation for future development.

As far as the back-end is concerned, the most challenging part was to prepare the testing environment that was easy to manage, yet integrated with all important services.

This required us to either:

  • Use a testnet (eg. Sepolia) that was supported by external API, like NFT API from Alchemy (fetching tokens owned by address), and OpenSea (token valuation).
  • Or create a local blockchain and mock the requests.

Another significant hurdle involved integrating the blockchain-based features. Direct communication between frontend, backend, and the blockchain often led to connectivity issues. Close collaboration between front-end developers and blockchain specialists was necessary to ensure seamless integration and functionality.

Additionally, data stored on both the backend and the blockchain required meticulous synchronization - a responsibility handled at the front end.

The Process

The Process

UX Design

Our client had very well thought-off user flows but the challenge we faced was understanding them and designing in a way that was intuitive to the users despite it being the product’s MVP version. The main UX goal was then to ascertain the sense of transaction security and trust between two parties, as well as its quick and easy completion.


Creating a test environment was the biggest challenge. Eventually, we used two environments. Testnet took a key part in integration tests, while a local blockchain allowed us to perform tons of tests and check the edge cases. The first solution was treated as a staging environment and the second as a development one.


We introduced a completely new registration system that improved the old user experience and introduced a slightly changed profile editing logic (not visible for the user) with some small tweaks for one of the screens styles. Based on our analysis of previous projects and considering the limited time we had for delivery, we decided to use “pure” ReactJS, supported by additional libraries, like redux- toolkit. Server side rendering was not so much needed here because we had a lot of dynamic data that needed to be loaded and rendered on demand.

Make sure that your product idea meets the market needs exactly.

Contact us



Web application

Thanks to the established user flows we could discern the most important functionalities and deliver wireframes for the application. Some of the main functionalities covered the adding of the offers to the marketplace by selecting NFTs as collaterals, a special button to quickly send desired loan terms to the owner of collateral, and the easy completion of the loan agreements when the terms were accepted.


We crafted a brand identity that feels friendly, modern, and stylish. Gradients, known for their innovative feel, were used alongside warm purples and pinks to create a welcoming and approachable atmosphere. To balance the inviting colors, we incorporated sharp and unique typography, adding a touch of decisiveness. Every element was meticulously crafted to achieve a distinct and visually striking look.

Business impact

  • The borrower can release liquidity from their NFT assets, not missing out on potential investment opportunities, while getting back NFT ownership by paying back the total loan principal and interest fee
  • In return, the lender (liquidity provider) receives a decent APR from the “Borrower.” If the borrower defaults, the lender will acquire ownership of the collateral NFT
  • Flexibility - providing the ability to extend the loan and early repayment options
  • Value-added supports - the platform provides functions like NFT valuations and trading statistics data serving as a reference for users’ decision-making
  • Enhanced efficiency - the interaction tool and user statistic chart function can greatly improve the matching efficiency
  • Multi-chains and multiple collateral support

Check other case studies


10Clouds AI Assistant

10Clouds Assistant is a powerful tool that can collect data from company handbooks, employee information, project details, and more to provide instant feedback. The process of creating this tool involved several stages, such as downloading data from the Confluence API and saving it to a PostgreSQL database. Next, we cleared the data of HTML tags and added it to Pinecone using Llama-index, a library for importing data. Our AI tool can quickly and accurately provide users with the information they need, making it an invaluable asset for any organization.

See more

In need of a tailor-made web3 solution for your business?

Contact us