Microsoft loves Python + AI

AI News


And we're showing it by launching a new step-by-step tutorial for building AI-powered applications using Python.


In fact, it's a nine-part Microsoft Reactor live series, complete with slides and code, so if you take it from start to finish, it collectively acts as a course.

Physically, everything is there. Assess AI quality and safety with embedded models, vision models, RAGs, tool calls, structured output, automated evaluation and red teaming. And on top of that we have Agentic AI and MCP. What else can I ask for?

This series, simply called “Python + AI”, consists of the following “lectures”:

  1. large language model
  2. vector embedding
  3. Search extension generation
  4. vision model
  5. structured output
  6. quality and safety
  7. tool call
  8. agent
  9. model context protocol

You will notice that the difficulty of the lessons increases. It starts with the basics of LLM and ends with agents and MCPs.

The lesson on “Vector Embedding” is particularly engaging as it explores the various vector embedding models that are the heavy lifting of the RAG pipeline.

The Invoking Tools, AI Agents, and MCP chapters work together, each setting the foundation for what follows. In particular, the last MCP chapter shows you how to use the Python FastMCP SDK to build a locally running MCP server and consume it from a chatbot such as GitHub Copilot.

Throughout the course, Python code is used to interact with LLM using common packages such as OpenAI SDK, LangChain, LangGraph, and Microsoft Agent Framework. To proceed with this procedure, you need a GitHub account to use the GitHub Models service, a provider of free models.

When it comes to the quality of the learning materials, it's top-notch, and you can be assured of that knowing that it was created by none other than Pamela Fox, a Microsoft cloud advocate with a solid background in creating educational materials for all kinds of Microsoft products. We recently checked out another great product of hers on “Microsoft's RAG Time”.

In conclusion, Python+AI is a great foundational series, a step-by-step guide to building AI-powered applications using Python. Recommended!

Detailed information

Level up your Python + AI skills with the complete series

Related articles

Microsoft's RAG Time

To stay informed about new articles in I Programmer, sign up for our weekly newsletter, subscribe to our RSS feed, and follow us on Facebook or Linkedin.

banner


picobook


comment

Or email your comments to comments@i-programmer.info.



Source link