Introduction

  • 3 minutes

Suppose you're a developer for Contoso Solutions, a leading provider of enterprise software and cloud-based services for businesses around the world. Contoso specializes in helping organizations modernize their infrastructure, automate workflows, and accelerate software delivery with innovative DevOps practices. With a focus on agility and reliability, Contoso is constantly exploring new technologies to streamline development operations and improve deployment processes.

To further enhance their internal DevOps pipelines, Contoso is piloting the use of cutting-edge AI to assist developers and operations teams. They envision an AI assistant that can help automate common DevOps tasks—like building environments, creating branches, deploying applications, and reviewing logs—making their workflows more efficient and reducing manual errors.

You're tasked with creating a proof of concept for this DevOps Assistant, demonstrating its ability to interact with users in natural language, understand DevOps-related requests, and perform basic operations. To power the assistant's intelligence and task execution, you choose to use the Semantic Kernel SDK.

This module guides you through the steps required to develop an AI DevOps assistant using the Semantic Kernel SDK. By the end of this module, you will complete a small assistant application that can interpret user requests, simulate DevOps actions with mock functions, and showcase how AI can enhance software development operations.

Note

This is a guided project module where you complete an end-to-end project by following step-by-step instructions. 

Learning objectives

In this module, you practice how to:

  • Create plugins for the Semantic Kernel.
  • Create prompts to elicit the best responses from the large language model.
  • Use the chat history to guide generated responses.
  • Automatically invoke the correct plugins to complete tasks.
  • Create function filters.

Prerequisites

  • Experience programming in C#.
  • Visual Studio Code IDE installed.
  • Familiarity with Azure and the Azure portal.
  • Access to Azure OpenAI Services.
  • Experience creating plugins using the Semantic Kernel SDK.
  • Experience using Semantic Kernel auto function calling.