Topics

• Building AI that automatically investigates technical outages.
• Evaluation metrics and scorecards for AI quality.
• LLM-based evaluation approaches.
• Team-based systematic improvement.
• The new AI Engineer role at incident.io.
• Build vs buy for evaluation tools.

Context

First major podcast appearance of the year. Came after Q1 push to establish thought leadership via Pragmatic Engineer + Building with AI microsite.