ai-cachara

About the Project:

This FullStack application serves as a training for AI concepts and structure within the .NET environment. It also serves as a training ground for frontend development using typescript, react and Next.js.

In the end of this project, the application should:

1. Classify document píctures quality(good, medium, poor);
2. Extract data from the documents with:
   1. LLMs APIs (semantic Kernel);
   2. Custom AI Local Model, with training data (ML.NET).
3. Store the pictures and extrated data into a non-relational database;
4. Have a Next.js Interface in which you can upload files for classification;

Technologies

1. .NET 9;
2. ML.NET and Semantic Kernel;
3. Non-relational database;
4. React in Next.js;
5. Typescript;
6. Consistent coding style with .editorconfig file

Project Structure

1. Backend

AICachara.Core

Application Services, Entities, Infrastructure and ML related code

AICachara.API

Endpoints for file upload, testing and providing content for the UI

2. Frontend

ai-cachara-ui

Next.js application that provides the User Interface of this application.