How Machine Learning Fails with Megan Robertson

About Show #1040

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

Recorded May 7, 2026

 

Megan Robertson is the founder of MegRob Data Science and Analytics, a data science and machine learning consulting firm specializing in helping organizations translate complex technical capabilities into real business impact. Prior to starting her own venture, she spent six years at a Fortune 500 company working at the intersection of technical and non-technical teams, where she designed and delivered ML and analytics solutions to support strategic business goals. She is deeply committed to expanding access to the field and is passionate about helping individuals from underrepresented communities break into data science and machine learning. When not running her business Megan enjoys skiing and climbing in her home state's mountains, traveling to new countries, and photographing the places she explores.
 

Show Comments

blog comments powered by Disqus