What changes when 300 dbt users at Virgin Media O2 move to Fusion and dbt State
Virgin Media O2 has 42 million customers, 300 dbt users, 100 dbt projects, and spends millions of dollars in compute annually. At that scale, inefficiency isn't a nuisance. It's a cost.
This session shows what happened when we adopted the dbt Fusion engine across our single-tenant environment. Wait times fell. Compute spend dropped. And we broke our reliance on slow CI/CD cycles. We'll show the before-and-after with real numbers, quantifying the measurable gains we achieved in both cost and speed. We'll also cover how implementing static analysis raised quality, consistency, and trust across 100 dbt projects.
We'll close with what we're building on this optimized foundation: AI, dbt skills, and agentic workflows. This is about more than tooling. It's about unlocking speed, efficiency, and intelligence at scale.
Check out more sessions
- Breakout session
AI-augmented analytics engineering at DocuSign: from req to release-ready with Fusion, dbt MCP, and...
Bishal Gupta / DocusignView session - Lightning talk
What AI agents get wrong, and how semantics fixes it
Sonny Rivera / ThoughtSpotView session - Breakout session
Don't ship what you can't measure: Evaluation-driven development for AI analytics agents
Opeyemi Fabiyi / Data CultureView session
