Standardizing AI Context: the Model Context Protocol

Standardizing AI Context: the Model Context Protocol

The Model Context Protocol (MCP) server acts as a centralized context management backend for advanced AI applications.

Unlike traditional session or prompt-based approaches, the MCP server manages a persistent, structured context that can be queried, updated and synchronized across distributed AI components and user sessions. By providing a standard API for serializing, retrieving and sharing context, the MCP server enables seamless interoperability between different models, services and clients. This is particularly important in multi-agent systems, long-running workflows or environments where state continuity and context portability are required.

Read Blog Post
When to use records in C#

When to use records in C#

For several years now, C# has supported Records as a new “type”. Initially only for classes (which is why the class is optional for a record class), later also for structs. Records are specially designed for use as data containers aka DTOs (please dont use DTO in your class name!). They are immutable and have special semantics for equality.

Read Blog Post