Demo Dive Series | Part 3 of 3

Putting AI to Work in Production: Playbooks and Built-In Agents for Governed Data Workflows

Encode senior-engineer judgment into workflows that run on a schedule, with a person still signing off

10 November 2026 | 10 a.m. CET / 9 a.m. GMT

Run the workflows that depend on your most senior engineer, even in the weeks that engineer is away.

A skill handles one task. A nine-phase governance rollout spanning eight to 12 weeks isn’t a single task, and neither is triaging 38 simultaneous alerts. Playbooks encode the whole workflow: who decides, who executes, and what happens when the happy path breaks. The AI follows the structure rather than inventing it, so work runs with your senior engineer’s judgment even when they aren’t the one running it. In this final session, Mikkel Dengsøe (Director of Data & AI Strategy, Coalesce) shows two production playbooks built on Coalesce MCPs: one where a person drives a Catalog governance rollout with AI executing each phase, and one where AI runs incident triage end to end with a person reviewing the result. He also shows where built-in agents fit instead, with Coalesce Copilot in the Transform build path and Scout, the always-on data SRE in Coalesce Quality, and how to decide between a prompt, a skill, a playbook, and a built-in agent for any given workflow.

You’ll learn how to:

  • Structure the human-AI handoff — compare two opposite patterns: a person leads and AI assists phase by phase, or AI drives the investigation and a person signs off.
  • Encode knowledge that would otherwise walk out the door — build anti-patterns, edge-case logic, and domain gotchas into playbooks so every run carries senior-engineer judgment.
  • Make every run auditable — replace subjective sign-offs with measurable thresholds and exit criteria the AI can verify itself, and recognize when a built-in agent is the better fit.

This session is broadcast for European audiences, and Mikkel joins live in the chat to take your questions.

Part 3 of a three-part series. See the full series

Speaker

Mikkel Dengsoe
Mikkel Dengsøe
Director of Data & AI Strategy
Coalesce