AI Development in .NET - Microsoft.Extensions.AI

Slide 1

Slide 1 text

AI Development in .NET Mert Metin #dotnetkonf25 Microsoft.Extensions.AI

Slide 2

Slide 2 text

Interests Microsoft / .NET software architecture, secure software, clean code infrastructure, system design Senior Software Engineer Blogger, speaker MERT METİN QR to reach me

Slide 3

Slide 3 text

Agenda Basic Concepts of GenAI Exploring the .NET Package - Microsoft.Extensions.AI

Slide 4

Slide 4 text

Brief Recap of the GenAI Concepts Generative AI, LLM, Prompt, Roles, RAG ,Agent, Token, Embedding, Tools

Slide 5

Slide 5 text

Generative AI Subset of AI Creates new content such as text, code, audio, image based on user prompts and learned data.

Slide 6

Slide 6 text

Large Language Model LLMs are optimized for complex tasks, trained on huge datasets, and capable of understanding and interpreting human language.

Slide 7

Slide 7 text

Prompt An input that is used to communicate with the model to generate desired output. Prompt Engineering is the practice of designing effective prompts. Prompts must be; clear, specific, goal oriented, have contextual information

Slide 8

Slide 8 text

System Specify “How will the model behave?” Roles User Prompt and query by user. Asistant Model responds using user’s prompt. It is used to ensure focus on a specific task or behavior pattern.

Slide 9

Slide 9 text

Agents They are capable of reasoning, planning, and interacting with their environment, using generative AI to perform tasks autonomously.

Slide 10

Slide 10 text

Tool Functions provided to LLMs must have clear objectives. Tools enable agents to perform actions such as web search, retrieving information from external sources and API calls.

Slide 11

Slide 11 text

Token Meaningful unit processed by LLM. Token is not only a word. tokenized by gpt-4o

Slide 12

Slide 12 text

Embedding - Vector Database Embedding is an operation that converts text, images into numerical data (vectors) that machines can understand. A vector database stores numerical data as vectors, enabling efficient similarity search

Slide 13

Slide 13 text

Retrieval Augmented Generation (RAG) Enhances LLMs with external data sources (databases, documents, APIs). Improves response relevance and accuracy.

Slide 14

Slide 14 text

Exploring the .NET Package Microsoft.Extensions.AI Provides utilities for working with generative AI components.

Slide 15

Slide 15 text

Why Support for switching between different LLM providers Manageable, Simple, and Rapid Development Provide implementations for dependency injection, caching, telemetry and other cross-cutting concerns. Implement chat and embedding features using any LLM provider.

Slide 16

Slide 16 text

IChatClient Connecting to LLM providers such as Open AI, Ollama, Azure Open AI

Slide 17

Slide 17 text

Microsoft.Extensions.AI.OpenAI LLM provider for OpenAI Working With Model Providers

Slide 18

Slide 18 text

Microsoft.Extensions.AI.Ollama LLM provider for Ollama Useful for working on local machine Working With Model Providers Note: This package is deprecated and the OllamaSharp package is recommended.

Slide 19

Slide 19 text

OllamaSharp The eaisest way to use the Ollama in .NET Working With Model Providers

Slide 20

Slide 20 text

GetResponseAsync Sends a user chat text message and returns the response messages. Return type is ChatResponse which represents the response to a chat request.

Slide 21

Slide 21 text

GetStreamingResponseAsync Sends chat messages and streams the response. Return type is IAsyncEnumerable which provides a stream of updated chat response.

Slide 22

Slide 22 text

Case - Sentiment Analysis Customer’s order reviews of the restaurant

Slide 23

Slide 23 text

Output - GetResponseAsync

Slide 24

Slide 24 text

Output - GetStreamingResponseAsync

Slide 25

Slide 25 text

Conversation Application needs to keep conversation context. So, send conversation history to the model using roles. Assistant role is set by AddMessages. Case - git commit generator

Slide 26

Slide 26 text

History and Roles ChatMessage list often represents the history of whole chat messages that are part of the conversation. Chat result

Slide 27

Slide 27 text

Tool Calling UseFunctionInvocation - set method name as delegate type

Slide 28

Slide 28 text

Tool

Slide 29

Slide 29 text

Output Usage Informations Response

Slide 30

Slide 30 text

Embeddings GenerateEmbeddingAsync has more detailed information about embeddings and Vector collection is also included. GenerateEmbeddingVectorAsync returns the vector collection.