Listen to this Post
The easiest way to start using AI in .NET is by leveraging the Microsoft.Extensions.AI library alongside Large Language Models (LLMs). In this guide, we’ll explore how to implement semantic search in .NET using Ollama and the .NET AI library in fewer than 50 lines of code.
🔗 Reference Video: How to Create Semantic Search with .NET & Ollama
You Should Know: Key Steps & Code
1. Setting Up the Environment
Install the required NuGet packages:
dotnet add package Microsoft.Extensions.AI dotnet add package OllamaSharp
2. Configure Ollama for Local LLM
Run Ollama locally (requires Docker):
docker run -d -p 11434:11434 ollama/ollama ollama pull llama2 Download a model
3. Implementing Semantic Search in .NET
using Microsoft.Extensions.AI;
using OllamaSharp;
var builder = WebApplication.CreateBuilder(args);
// Add AI services
builder.Services.AddAIServices(ai =>
{
ai.UseOllama("http://localhost:11434", "llama2");
});
var app = builder.Build();
app.MapGet("/search", async (IAIService aiService, string query) =>
{
var embeddings = await aiService.GenerateEmbeddingsAsync(query);
// Compare with stored vectors (e.g., from a DB)
return Results.Ok($"Search results for: {query}");
});
app.Run();
4. Testing the Semantic Search
curl "http://localhost:5000/search?query=What%20is%20.NET%20AI?"
5. Advanced: Storing & Retrieving Embeddings
Use Azure AI Search or PostgreSQL with pgvector:
CREATE TABLE documents ( id SERIAL PRIMARY KEY, content TEXT, embedding VECTOR(1536) );
What Undercode Say
Semantic search revolutionizes traditional keyword-based search by understanding context. Here are some related commands and tools to deepen your knowledge:
Linux & IT Commands
Monitor Ollama logs docker logs -f ollama_container Check GPU usage (if using CUDA) nvidia-smi Install .NET on Linux sudo apt-get install -y dotnet-sdk-8.0
Windows & .NET
List installed .NET SDKs dotnet --list-sdks Benchmark your AI service dotnet counters monitor Microsoft.AspNetCore.Hosting
AI & Cloud Integrations
- Azure AI Studio: https://azure.microsoft.com/en-us/products/ai-studio
- Hugging Face Models: Integrate with `transformers` in .NET.
Expected Output:
A functional semantic search API in .NET, scalable with vector databases, and optimized for low-latency AI inferences.
🔗 Further Reading:
References:
Reported By: Djokic Stefan – Hackers Feeds
Extra Hub: Undercode MoN
Basic Verification: Pass ✅
