From the course: Microsoft Azure AI Engineer Associate (AI-102) Cert Prep by Microsoft Press
Join today to access over 24,900 courses taught by industry experts.
Use Semantic Kernel and Autogen in agent workflows - Azure Tutorial
From the course: Microsoft Azure AI Engineer Associate (AI-102) Cert Prep by Microsoft Press
Use Semantic Kernel and Autogen in agent workflows
- [Instructor] Well, here we come back to the concept of agent versus assistant. Here's the distinction again, at least these are aspects of that distinction. An assistant is one conversational thread. It's a special purpose project domain specific bot that has its own system prompt, it has some RAG capabilities, but it's not agentic. A true agentic agent is able to work in a parallel fashion, can support persistence, tools, persistent memory, goals, much more autonomous AI. And as we've seen in the previous lesson, the agent platform is under Azure AI agent service. What about reflection and tool use? What does this mean? Well, reflection is different from reasoning, agents that think and retry. For example, you might create a custom agent for a certain kind of software development and when the agent tries a routine, let's say, maybe there was a syntax error or some other bug. Reflection means that the agent is able to spot that, oh, and figure out what happened and autocorrect it on…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.