This is an AI chatbot application created with Next.js and Nest. The chatbot history is stored in a local sqlite database. All responses are generated using the OpenAI API and enriched with a use case specific prompt.

Check out the demo video to see it in action.

Running the application

First, you need to provide the OpenAI token. Create an .env file inside the backend folder and add the token as following:

OPEN_AI_API_KEY=sk-proj-RwCkh...

Then run the backend application to start the API:

$ cd backend
$ yarn install
$ yarn run start:dev

It should now run on http://localhost:3001. The backend will create the database.sqlite file if it doesn't exist.

Second, run the development frontend application:

$ cd frontend
$ yarn install
$ yarn dev

Open http://localhost:3000 with your browser to see the webapp.

General comments

All the requirements are met beside chat response streaming. I skipped it due to time constraints.

Note that this application is a proof of concept. It is built with good principles in mind but the code can be made much cleaner. Same applies for styling and UX.

A few things that should be worked on if the app would be deployed to users:

• database (it might be better to migrate to a more scalable solution, like a cloud database)
• authorization (user id is hardcoded on the backend side, it could be properly done with an OAuth token)
• error handling
• reusing models between frontend and backend
• loading states
• tests