Beyond a Reasonable dbt

Trustworthy pipelines for AI agents. The same argument comes in two formats: an eleven-slide briefing for decision makers, and a full technical walkthrough for the people who will build it.

For decision makers

Executive Briefing

Eleven slides on the cost of click-ops agents, what managing the lifecycle as code changes, and the ≥95% correctness bar to ship. Runs full screen with arrow keys and prints one slide per page for a PDF handout. No implementation detail.

Slide deck11 slides, about 15 minutes

Present →

For builders

Technical Deep Dive

The opening statement, a five-phase build runbook with the exact dbt commands, and three exhibits on semantic views, agent specs, and evaluations. Tabbed, printable, and safe to hand out as a link.

Technical deep dive5 sections, tabbed

Open the case →

What both versions argue

An AI agent is only as trustworthy as the data and the process behind it. Yet most agents are assembled by click-ops in a UI: no version history, no peer review, no tests, and no reliable way to promote the same thing from dev to prod. When the agent gives a wrong answer, there is nothing to diff and nothing to roll back. Managing the full lifecycle as version-controlled code with dbt Projects on Snowflake fixes that.

About this deck

Built from the cortex-agents-dbt-project-template (see dbt Projects on Snowflake). The walkthrough mirrors the project's WORKING-SESSION.md runbook; the exhibits draw on its README.md best practices.