Anonymise a document before handing it to an AI model, from .NET
Eleven-step procedure: Azure AI Language finds the personal data, the program swaps it for placeholders, only the masked text reaches the model, and the real data goes back in at the end. With the limits measured in practice.
Vector search with Azure AI Search from .NET
Eleven-step procedure: a search service on the free tier, keyless access, an embedding model, an index with a vector field created from .NET, the documents uploaded, a search by similarity, then delete the resources.
Deploy a model on Microsoft Foundry and call it from .NET
Twelve-step procedure: a Microsoft Foundry project, a gpt-5-mini deployment, a test from the playground, a call from a .NET console application with Entra ID, then delete the resources.