Hello Reader,

The next episode of the Microsoft Cloud IT Pro podcast has been released!

A few months ago, Ben and Scott started experimenting with LLM Wikis. But what happens when you stop treating an AI knowledge base as an experiment and start relying on it every day?

In Episode 438: Taking Your WorkOS and LLM Wiki to the Next Level, they revisit those systems and compare what happened after months of real-world use.

The result is something much closer to a personalized WorkOS—a system that brings together tasks, meetings, research, communications, knowledge, and recurring workflows.

In this episode, Ben and Scott discuss:

• How an LLM Wiki can evolve into a personalized WorkOS
• Daily skills for prioritization, meeting prep, meeting capture, and end-of-day reviews
• Moving from on-demand skills to scheduled workflows, loops, and deterministic Python scripts
• Using Obsidian, Markdown, Asana, and Microsoft 365 as different parts of the system
• Building dashboards and command centers for accessing workflows
• Keeping governed documents in Microsoft 365 while still making AI useful
• Filtering Teams activity to separate useful information from noise
• What these personalized AI systems actually cost to run each month

They also explore an important shift in how these systems mature: not everything needs to remain an AI-driven workflow. Once a process becomes predictable, parts of it can move into deterministic scripts that are easier to call from different interfaces and more consistent to run.

If you listened to Episode 429 on getting started with LLM Wikis, this episode is the next chapter—what happens after you build the system, live with it, and start refining it around the way you actually work.

Like what you hear and want to support the show? Check out our membership options.​

Show Notes

• ​A pattern for building personal knowledge bases using LLMs
• ​LLM Wiki v2: extending the pattern with lessons from production
• ​Agent Skills open standard
• ​About agent skills for GitHub Copilot
• ​Use Agent Skills in Visual Studio Code
• ​Building effective agents
• ​Conductor: Deterministic orchestration for multi-agent AI workflows
• ​Obsidian Web Clipper

About the sponsors

​Nasuni helps enterprises manage, protect, and activate unstructured file data so teams can work smarter, securely, and at scale.

Would you like to become the irreplaceable Microsoft 365 resource for your organization? Let us know!​

View today's episode

​https://www.msclouditpropodcast.com/episode438​

​​​​​​