Case study

SharpChat.ai

SaaS platform for building AI Chatbots for customer support.

SharpChat.ai
View project online

Description

SharpChat.ai is a Software as a Service (SaaS) that allows the user to create, manage and customize an AI chatbot to handle customer support requests.

The AI ​​chatbot can be installed on the company website or can be shared with a link.

The AI ​​chatbot leverages OpenAI LLM models such as GPT3.5 and GPT4 to create instant answers to customer questions.

The user can add documents, websites, web pages or text blocks to the chatbot's memory so that the chatbot can remember the information.

The user can apply corrections to improve the quality of answers.

The user as a dashboard where he can manage the entire chatbot by himself.

Technologies used in this project

Programming Languages:

C# C# Javascript Javascript SQL SQL HTML HTML CSS CSS

Frameworks, Libraries and More:

.NET .NET ASP.NET Core ASP.NET Core EF Core EF Core SQL Server SQL Server Azure Azure Bootstrap Bootstrap Tailwindcss Tailwindcss Svelte Svelte OpenAI OpenAI Qdrant Qdrant Stripe Stripe

Software architecture

Frontend

Website

Built using ASP.NET Core Razor Pages + Bootstrap and the website is hosted on Azure App Service

User Dashboard

Built using ASP.NET Core Razor Pages + Bootstrap and alpine.js for rich UI interactivity.
Leverage Entity Framework Core compatibility to build fast UIs for CRUD operations.

Chatbot Interface

Built using Svelte via Vite + Tailwindcss, the application is compiled into a single HTML page and is hosted via Firebase Hosting.

Chatbot Widget

Builded using Svelte via Vite + Tailwindcss.
The application is compiled into two files, a javascript and css file that can be imported into any website to show an interactive widget.
The widget javascript and css files are hosted on a CDN via Azure Storage.

Backend

API and Logic

Builded using ASP.NET Core and uses REST endpoints to allow the Chatbot interface and Chatbot widget to communicate with it.

Database

The database is built and managed using Entity Framework Core as an Object-Relational Mapper.

User Identity

User identity and authentication process are implemented using built-in ASP.NET Core Identity Framework.
These are the implemented operations: Email Sign-In, Google Sign-In, Forgot Password, Email Confirmation, GDPR Data Download, Account Delete.

External Services

The applications use these three external services:
  • OpenAI: To generate AI responses
  • Qdrant: To save and retrieve user information used to personalize the chatbot
  • Stripe: To automatically manage user sign-ups

Free Consultation

Discover how to improve your business with generative AI and/or custom software.

Includes:
  • Business needs analysis
  • Software design
  • ROI calculation
  • Quote with price and timing
Free Consultation
Quote