- Overview
- Features
- Prerequisites
- Installation
- Usage
- Project Structure
- Console output
- Resources on Semantic Kernel and AI Integration
- License
- Keywords
- Key Topics and Challenges addressed by this project
- Problem or Opportunity Statement
- Who is this for
The Virtual Customer Success Manager (VCSM) is a C# .NET project that leverages Semantic Kernel and Azure OpenAI to deliver personalized onboarding, proactive support, and tailored recommendations for Microsoft products. The application dynamically responds to user queries using AI-powered agents and plugins.
- Personalized Onboarding: Provides step-by-step guidance for setting up and using Microsoft products, using Semantic Kernel Plugins.
- Proactive Support: Implements a multi-agent system to proactively address user issues, ensuring system stability and optimal performance.
- Tailored Recommendations: Offers personalized suggestions and tips based on user behavior and preferences, retrieved via Semantic Kernel's Vector Store.
- .NET SDK (version 8.0 or later)
- Azure OpenAI credentials
-
Clone the repository
git clone <repository-url> cd <repository-folder>
-
Set up User Secrets The project uses Azure OpenAI for processing natural language queries. To configure this securely, use
dotnet user-secretsto store the API key and related configuration.dotnet user-secrets set "AzureOpenAI:DeploymentName" gpt-35-turbo-16k dotnet user-secrets set "AzureOpenAI:Endpoint" "https://<deployment-name>.openai.azure.com/" dotnet user-secrets set "AzureOpenAI:ApiKey" <insert-key>
-
Install Required Packages Install the necessary NuGet packages for the project.
dotnet add package Microsoft.SemanticKernel # version 1.18.2-alpha dotnet add package Microsoft.SemanticKernel.Agents # version 1.18.2-alpha dotnet add package Microsoft.SemanticKernel.ChatCompletion dotnet add package Xunit
-
Run the Project Execute the
Program.csfile, and follow the prompts to select a desired feature.dotnet run
-
Pick an Option When running the project, you will be prompted to select from the available functionalities:
Please choose an option (type 'end' to exit): 1. Personalized Onboarding using Semantic Kernel Plugins 2. Proactive Support using Multi-Agents 3. Tailored Recommendations using Vector Store Enter input (1, 2, 3, end to exit):- Option 1: Personalized onboarding walkthrough using Semantic Kernel Plugins.
- Option 2: Proactive support with issue resolution powered by multi-agent coordination.
- Option 3: Tailored recommendations based on user data stored in the Semantic Kernel Vector Store.
- InternalUtilities: Contains helper utilities for resource handling and configuration.
- Resources: Contains documents such as Teams onboarding guides and recommendations.
- _vcsm: Core VCSM logic split into distinct features like onboarding, support, and recommendations.
-
Personalized Onboarding using Semantic Kernel Plugins:
-
Proactive Support using Multi-Agents:
-
Tailored Recommendations using Vector Store:
- Microsoft Semantic Kernel
The Semantic Kernel is an open-source SDK that enables integration of AI models with other systems. It is the core technology used in this project to handle plugins, automatic function calling, and multi-agent orchestration. - Azure OpenAI Service
Azure OpenAI provides powerful language models, such as GPT, to create AI-driven, natural language understanding and response generation. It is used to provide the intelligent responses, tips, and personalized interactions in the VCSM project. - Semantic Kernel Dev Blog
Regular updates and insights on Semantic Kernel from Microsoft's development team. - Develop AI Agents using Azure OpenAI and Semantic Kernel SDK
Microsoft Learn training path that guides developers through building AI agents using Azure OpenAI and Semantic Kernel. - Introducing Agents in Semantic Kernel
Blog post detailing the use of agents in Semantic Kernel, key to building multi-agent systems like the one used in this project.
This project is licensed under the MIT License.
AI, virtual assistant, customer success, Microsoft products, onboarding, usage tips, feature recommendations, customer support, personalized advice, business value
- Customer Experience
- Operational Excellence
- AI Transformation
Customers struggle to fully use Microsoft products, leading to missed opportunities. An AI assistant offers proactive, personalized support to enhance usage and drive long-term success.
- Microsoft Product Users
- IT Support Teams
- Customer Success Managers
- Business Decision Makers
- Technology Enthusiasts