This project is a Proof of Concept (POC) aimed at creating a chatbot capable of providing knowledge in any format, including URLs, PDF files, TXT files, etc. In this instance, we focus on utilizing PDFs as the primary source of knowledge.
MindsDB is an open-source framework that allows you to easily create and deploy machine learning models and AI-powered applications. It simplifies the process of building intelligent systems by providing a user-friendly interface and seamless integration with various data sources. By leveraging MindsDB, we can efficiently build our chatbot using its powerful machine learning capabilities.
Next.js is a popular React framework that enables developers to build server-rendered and statically generated web applications. It is known for its performance and optimization features, making it a perfect choice for our project, as it allows for rapid development and deployment of the frontend components of our chatbot.
- Bun Runtime: Install Bun by following the instructions on the official website: Bun.sh.
- Docker Desktop: We will use Docker to run the MindsDB backend in a container. Ensure you have Docker Desktop installed.
-
Create an OpenAI API key by visiting this link: OpenAI API Keys.
-
Add your OpenAI API key in the
frontend/.env.localfile. -
Install dependencies in the root directory as well as in the
frontendfolder by executing the following commands:bun i cd frontend && bun i cd ..
To run the project, you can use the following commands:
-
Development:
bun run development
-
Production:
bun run production
If you need to reset the backend (i.e., clear files, models, agents, knowledge bases, etc.), you can use the following command:
bun run clean-backendWe welcome contributions and suggestions for improvements! Please feel free to open issues or submit pull requests to enhance the functionality of this chatbot POC.
This project is licensed under the MIT License.