Tagged: Prompt Engineering
8 posts · All tags
-
Same Prompt, Two Reports: Give Desktop Bridge a Spec Before It Builds
October 2, 2026
Give Desktop Bridge the same prompt twice and you get two different reports. The three report skills and a real kickoff page are how you get the same one.
Read More → -
A Data Agent Should Be a Sub-Agent
September 23, 2026
A Fabric data agent works best as a sub-agent with one job that Copilot, Foundry, or Copilot Studio calls. Here's how I write those instructions.
Read More → -
A skill.md Is Not Wisdom. So Make the Agent Grill You.
September 21, 2026
Your skill file carries instructions, never the judgment that says this is six hours not two. Matt Pocock's grill-me is how I get the judgment back in.
Read More → -
Build the Thing That Creates the Thing
September 16, 2026
Stop prompting for the same notebook edit. Build the executor: an agent, its skills, and the context it reads, in a harness that reaches Fabric and Power BI.
Read More → -
Design the Report From the Meeting You Already Had
August 19, 2026
Agents can design Power BI reports now. What makes them good is a locked brief from your stakeholder meeting, plus commands and the Desktop Bridge loop.
Read More → -
Don't Let Your Agent Touch Fabric Until It Reports Back
August 17, 2026
An agent's bad PySpark doesn't get a red squiggly line. It gets you data. The read-only survey I open every Fabric MCP session with, and the asserts that gate everything after.
Read More → -
Stop Re-Prompting: Your Second Brain Should Write Your Agent's Instructions
July 20, 2026
Stop re-prompting your AI agent. Build a two-harness workflow where your second brain writes the agent's instructions and MCP executes in Power BI and Fabric.
Read More → -
Unlocking AI for Power BI & Business Intelligence Skills
July 17, 2024
I have been lucky enough to be in the Business Intelligence space for the past 9 years with a focus on Power BI and the amazing application it has for organizations.
Read More →