Custom AI-powered chat application 

Country

USA

Domain

Communication & Collaboration

Service

SaaS services

Cooperation time

10/2023 – ongoing

Team size

4 people

Technologies:

.NET, C#, Blazor, SQL Server, Microsoft Azure DevOps, OpenAI API, Azure OpenAI SDK, Semantic Kernel, RAG, Azure SignalR, WebRTC

Background

Our client, operating in the USA, has undertaken a strategic initiative to revolutionize how people interact and collaborate in virtual space. The ask was an MVP of AI-powered chat application (web-based for a start) which will integrate with artificial intelligence capabilities. The key feature of this solution is the inclusion of a sophisticated conversational AI and copilot designed to formulate and send messages intelligently on behalf of the user.

AI powered Chat Application

Challenges

One challenge in developing this application was the strategic imperative to effectively empower users to manage their knowledge base within the platform. The client aimed to enable users to define specific groups with whom the AI assistant could share designated portions of their knowledge base. This required a thoughtful and granular approach to permissions, ensuring that users could precisely control the dissemination of information based on the context and relevance to different groups.

AI Powered Customization

Features

  • Private & group chats
  • Video calls
  • Personal knowledge base management, customized security controls
  • AI assistant with RAG pipeline
  • AI Copilot

AI Assistant

Solution

Reenbit has quickly set up the Microsoft Azure-based architecture required to host web applications, store data, and manage communication channels.
The next step was to select the proper LLM provider that would cover application requirements with data security in the top place. After careful consideration and analysis, we integrated OpenAI’s GPT-3 through the Open API Assistants API. Since we needed to configure AI instructions per communication group granularly, the OpenAI’s concept of assistants, threads and messages was the best candidate for implementing the requirement. Moreover, the availability of the Open API Assistants API provided a convenient and streamlined integration process.

Intelligent Communication System

Outcome

The MVP implementation of an AI-driven custom chat application resulted in the following:

  • User satisfaction increased
  • Time spent on customer queries was reduced, leading to a 20% increase in productivity.
  • The client’s innovative approach positioned them as pioneers in the communication industry and laid a robust foundation for future business expansion.
  • Brand perception improved, leading to a 30% increase in user engagement.

The next versions will include advanced AI features to improve the user experience and operational efficiency.

Your browser does not support the Canvas element.

Tell us about your challenge!

Use the contact form and we’ll get back to you shortly.

    Our marketing team will store your data to get in touch with you regarding your request. For more information, please inspect our privacy policy.

    thanks!

    We'll get in touch soon!

    contact us