Member-only story
Semantic Kernel 101
Part 2: Understanding Multi-Agents frameworks
13 min readJan 5, 2025
In my previous article, we introduced the lightweight framework of Semantic Kernel (SK) and how to build one single AI Agent. In this article, we will cover how to create multi-agents system, with a focus on the different selection patterns.
What is Multi-Agency?
In the second half of the 2024, AI Agent has become the new buzzword whenever talking about LLM-powered applications. This is due to the fact that they proved to be extremely efficient at solving tasks and automating business processes. At their core, AI Agents consists of the following essential elements:
- An LLM: This serves as the reasoning engine, driving the agent’s ability to analyze and generate responses, as well as deciding when to invoke a specific tool to accomplish user’s query. It can be seen as the “brain” of the agent, orchestrating user’s requests and the backend of the Agent itself.
- A Toolkit: This is the set of resources the agent can use to interact with its environment. For instance, providing access to the Web Search qualifies as a tool for the Agent. The way the Agent decides which tool to use and when is determined by the LLM reasoning capabilities, as well as the planning strategy induced in the system message.