Qartora

Overview

An chat-like application that uses an AI agent to fetch, filter, and display products which are pulled from Shopify. The application has two entry points: WhatsApp and Instagram. Based on this, the application follows their respective designs, determined on the channel of entry.

The application other than the AI chat, contains a cart and creation of an order that is sent and processed through Shopify.

What I did

I started on this project by implementing the frontend using React with shadcn, while being aware of the work done in the backend which used NestJS. After implementing the reusable UI components and to follow the given design, I ensured that we had mocked data being pulled in a way that required minimal changes once the backend was done.

I was then working on the backend, since I have already prepared the mock to be ready for the connection with real data, I was tasked to connect the API data with the frontend. Afterwards, I started preparing the application to support multi-tenancy on a single deployment by using slugs in the path. While working on this, it was decided from the clients side that we would actually use a subdomain approach rather than the slugs, but we would still use a single DB.

Then I prepared the application to be able to support multiple channels (we started only with WhatsApp), so I made the UI to be configurable, as well as link generation based on the selected channel.

I was also assigned the 360 Dialog transfer from our application to the clients new backoffice app. Some UI / UX redesigning of the backoffice, etc.

Responsibilities

• Communicate with the client regarding desired functionalities of the system
• Present developed functionalities on a Sprint basis.
• Develop and test new functionalities, as well as maintain the system.
• Develop and execute plans for testing, and system validation.
• Ensure that the final product meets the client’s expectations.

Stack

Built with React, shadcn, NestJS, Next.js, PostgreSQL, Vercel SDK, TypeORM.