Joël and fellow thoughtbotter Jimmy Thigpen assess their AI workflows and question whether LLM partners really are as helpful as they claim to be.

Joël points out the blindspots AI can have when processing certain requests, they each share the ways they utilise AI into their workflow and pros and cons of doing so, as well as looking at some of the areas of improvement they would each like to see made to various AI agents in the future.

—

Thanks to our sponsors for this episode Judoscale - Autoscale the Right Way (check the link for your free gift!), and Scout Monitoring.

Your host for this episode has been thoughtbot’s own Joël Quenneville, and you can connect with this week’s guest Jimmy Thigpen over on LinkedIn.

If you would like to support the show, head over to our GitHub page, or check out our website.

Got a question or comment about the show? Why not write to our hosts: [email protected]

This has been a thoughtbot podcast.

© 2025 thoughtbot, inc.

Support The Bike Shed