What can go wrong with machine learning? While at NDC in Toronto, Richard chatted with Megan Robertson about her experience with machine learning projects, often using retail datasets, and where they can go wrong. Megan talks about getting clear expectations and metrics for projects, so you know when you succeed, but then digs into the specifics of problems in machine learning, such as overfitting on test data. Your results are only as good as the data you put in, so a lot of focus goes into building good sets, carefully developing the model with those sets, and using techniques like cross-validation to ensure the model is behaving appropriately. There's a lot that can go wrong, but the results with an effective model can be very powerful - it is worth the effort!

Links

• Cross Validate Model
• Megan's Website

Recorded May 7, 2026

How do you intelligently surface access to your database? While at NDC Toronto, Richard spoke with Jerry Nixon about Data API Builder, Microsoft's tool that enables data professionals using Microsoft databases, including SQL Server, Postgres, CosmosDB, and MySQL, to provide an API layer with security, schema extraction, and governance policies. You can expose the API as a REST interface, a GraphQL interface, and an MCP server! This is a powerful tool for providing controlled access to data while still allowing for ad-hoc access. The potential is huge - you need to check it out!

Links

• Data API Builder
• GraphQL

Recorded May 7, 2026

How can Microsoft Loop make your team more productive? Richard chats with Karinne Bessette about the role that Loop components can play in making meetings where the agenda is live, generating work items in Microsoft Planner, and keeping key information up to date. Karinne talks about how Loop components can be connected to any M365 document, including Outlook, Word, Excel, and OneNote, but only for members of the M365 tenant. Loop is a powerful tool for productivity within the organization!

Links

• Microsoft Loop
• Microsoft Planner
• Microsoft OneNote
• Power Automate
• Loop Components in OneNote
• Teams Polls
• Polls in Loop
• Loop Admin Policies

Recorded April 27, 2026