(AI Gen) Teaching AI to Read the Room: How We Built an Automated Incident Classifier
From Tilt Engineering by Tilt Finance
May 1, 2026 · 21 min · Episode 25
About this episode
This episode discusses the development of an automated incident classifier by Tilt, detailing the challenges faced and the technology used.
After 18 months and ~800 incidents, our Rootly classification fields were mostly empty, no one fills in six drop downs at 2am. This post walks through how Tilt built an automated incident classifier: a scheduled .NET pipeline that scrapes the Slack channel, sends the conversation to Claude, and writes back the cause, vendor, component, and team, while quietly backing off any field a human corrected.
Topics covered
- AI
- incident classification
- automation
- technology
- business
Keywords
- automated incident classifier
- Slack integration
- incident management
- AI technology
- Rootly
- Claude
- automation
Mentioned in this episode
Organizations: Tilt, Rootly
Products: Claude
Places: Slack
More episodes of Tilt Engineering
- (AI Gen) The Weak EM Will Become An NPC · June 9, 2026 · 20 min
- (AI Gen) Adventures in the Modern Monolith: How do you keep a secret at Tilt? · April 10, 2026 · 23 min
- (AI Gen) AI DevOps: Building Intelligent Incident Response Systems · February 13, 2026 · 14 min
- (AI Gen) Entropy Resistance · January 8, 2026 · 9 min
- (AI Gen) Set Your AI Up for Success: How BDD Tickets Drive Quality Outcomes · November 28, 2025 · 2 min
- (AI Gen) How We Built a Context-Aware Review Bot That Transformed Our Code Reviews · October 26, 2025 · 2 min
Explore listener stats, chart rankings, contacts and more on the Tilt Engineering podcast page.