Gen AI With Semantic Kernel for .NET Engineers

Create a console app to demonstrate basic prompts with Semantic Kernel. Define user and system messages, configure Azure OpenAI, and stream the JSON response via a kernel function.

Create a console chat loop by feeding an array of chat history strings as prompts to a semantic kernel, using a handlebars template, and print responses from Azure OpenAI.

Render prompts with the prompt template engine to chat with the kernel without executing them, then build a console app that uses time plugin and Azure OpenAI to render prompts.

Learn to process handlebars and liquid prompt templates in a .NET console app by configuring a kernel, rendering templates, and invoking the Azure OpenAI endpoint to obtain responses.

Build a .NET console app that uses a handlebars template prompt and a semantic kernel to process base64-encoded images, passing image data as arguments and printing an image description.

Explore prompting techniques for the semantic kernel, including instruction based prompting, few shot prompting, zero shot prompting, chain of thoughts, tree of thoughts, role based prompting, and adaptive prompting.

Explore semantic kernel, a lightweight middleware sdk that lets your dotnet, java, or python code integrate with AI models via plugins and filters, like Azure OpenAI, OpenAI, or Hugging Face.

Create a basic semantic kernel in a dotnet core app, configure Azure OpenAI with an API key, run prompts, and stream responses in JSON with max tokens and temperature.

Build a Q&A chat app with semantic kernel, a lightweight toolkit that connects Azure, OpenAI, and Olama via one NuGet package, routing prompts to the right AI model.

Build an AI function for a chat app with semantic kernel, featuring a city temperature plugin, kernel function integration, and streaming responses via the Azure OpenAI chat client.

Add plugins to a kernel by creating time information and a widget factory, enums, and a details class, then build and run the kernel with auto function selection.

Explore how to pass prompts to a semantic kernel using a yaml prompt file, leveraging semantic kernel and handlebars templates to generate topic- and length-constrained stories.

Inject semantic kernel into a dotnet core console app using dependency injection, build a service provider with plugins, and test kernel responses to prompts.

Create a console app to send chat prompts to a semantic kernel, configure an Azure endpoint API key and deployment name, and receive responses in JSON.

Learn how to implement a prompt filter with semantic kernel to block sensitive data like credit card numbers in a .NET console app, ensuring responsible ai behavior.

Learn to add and invoke plugins in a semantic kernel with dependency injection and without, and inspect prompt rendering, function names, and token usage in a console app.

Learn to integrate an open api plugin into a semantic kernel using dependency injection, inject a custom mechanic service, and run create and list repair operations with a swagger-driven API.

Ingest opentelemetry data from a semantic kernel into a .NET console app. Configure the kernel, enable trace and meter providers, and set up logging for observability.

Push OpenTelemetry data from a .NET console app to Azure Application Insights and Log Analytics, configure trace providers and a logger factory, and review prompts and kernel responses in traces.

Create a math tutor AI agent using a console app, configure Azure authentication, and run the agent to solve equations and print responses, with options for basic versus premium explanations.

Create a travel assistant agent that uses APIs to fetch weather and parks at a location, feeding tool outputs back to the user through a semantic kernel workflow.

Create a reasoning AI agent by deploying an oh three mini model in Azure and configuring a .NET project to query why the sky is blue using Azure OpenAI.

Create an AI agent with semantic kernel to upload a sales CSV, define the agent, and answer queries like total sales and top five countries that generated the most profits.

Build an AI agent with semantic kernel, add a menu plugin, connect via azure ai foundry, and stream responses live to answer restaurant menu questions.

Build an ai agent with chat completion via ChatCompletionAgent and auto function invocation using semantic kernel in a dotnet console app, featuring a Colonel host agent and a menu plugin.

Learn how to manage chat history in a semantic kernel agent by applying truncation, summarization, or token-based strategies to improve performance and memory efficiency while safeguarding privacy.

Implement a chat history summarization strategy in a .NET console app, using a summarization reducer with an agent and Azure AI foundry deployment to preserve context across 50–250 messages.

Learn to serialize and deserialize an AI agent chat with a group chat, kernel, and menu plugin in a .NET Semantic Kernel workflow.

Define a single kernel hosting two chat completion services, one with a good key and one with a bad key, and route invocations to the correct agent at runtime.

Learn to stream responses from a chat completion agent by building a console app, configuring Azure OpenAI settings, and printing streamed replies to chat history.

Learn to template an AI agent chat using semantic kernel, handlebars, and liquid formats, and implement in code with template configuration to switch formats seamlessly.

Build a two-agent workflow using Semantic Kernel where one agent generates files and a second agent summarizes the generated content with Azure OpenAI, group chat, and kernel setup.

Execute a multi-agent workflow where an analyst generates a chart and a summariser explains the chart output. Demonstrate image generation and summarization coordinated through Azure OpenAI tooling.

Learn how to reset a group chat across multiple agents, making them forget prior context to address privacy and security concerns while maintaining instruction constraints.

Serialize, deserialize, and clone a multi-agent group chat in a .NET console app, using translator and counter agents to manage turns and output.

Stream multi-agent responses in a group chat using semantic kernel for .NET, coordinating copywriter, art director, and reviewer with an approval termination strategy, leveraging azure openai endpoint.

Build a simple Rag application that uses an in-memory vector store to index movie data, generate embeddings, and perform vector searches with confidence scores.

Embed vector data into Azure AI Search and query it with drag, building a .NET console app to generate embeddings and query a vector store.

Discover how to build an in-memory semantic kernel and use retrieval augmented generation with a local llama model, leveraging kernel memory and embeddings for improved responses.

Learn to ingest data into an in-memory vector store and perform searches with semantic kernel and an embedding generation service in a .NET project.

Learn to search ingested data with semantic kernel by embedding text and querying Azure AI search as a vector store, using OpenAI text embedding service, Azure endpoints, and API keys.