Evals for AI Apps: Getting Started with promptfoo

Whether you're shipping a weekend chatbot or running an LLM feature in production, evals catch the regressions your tests miss.
A practical walkthrough of what to put in your first eval suite — and how to wire one up against a Python chatbot using promptfoo, quickly.

0:00 Intro
0:47 What evals actually catch
1:59 Why every role on the team should care
3:02 The concept — a regression suite for non-deterministic code
4:20 Demo
8:23 Bonus tips - Remember to add to CI
9:25 Bug fixes and production monitoring
10:16 Recap

Видео Evals for AI Apps: Getting Started with promptfoo канала Adeel Mansoor