Why AI needs a strong data foundation, and how to build it
Partner Capability: Business Value
September 23, 2026
Session 1: 3:00 PM BST / 10:00 AM EDT Session 2: 4:00 PM PDT / 9:00 AM AEST (next day)
Virtual
Justin Erickson
Kai Lee
What "AI-ready" data actually means: a fast overview of why most data stacks were built for human analysts, not AI agents asking questions around the clock, and where that gap shows up in practice
Context engineering in practice: how Fivetran + dbt establish a governed context layer so AI tools retrieve the right information instead of guessing, with real examples of what works and what doesn't
Integrating AI tools with trustworthy data: leveraging the dbt MCP server and agent skills so that a client's AI tools return accurate answers without burning excess tokens or compute
What this means for your business: how to turn this readiness work into an actual services conversation with clients who are already experimenting with AI on their own data, before they realize the foundation underneath it isn't ready
You're all set, [Name].
Registration closed.
Ever watched an AI tool confidently hand back a wrong answer because it was pulling from stale, disconnected data? That's not an AI problem; it's a data foundation problem. In this session, we get into the specifics of what actually makes data trustworthy for AI, with real examples of how to structure it and connect it with Fivetran + dbt so AI tools stop guessing.
Who should tune in:
- Solution architects and delivery consultants: if you're the one building or delivering AI use cases and designing client data architectures today, this is where the ground is shifting under AI initiatives. Come see what "AI-ready" actually requires before someone else figures it out first.
- Practice leads and account owners: your clients are already experimenting with AI on their own data. This is the technical grounding that turns that curiosity into a services conversation you can actually sell, and it gives your delivery teams a shared foundation of best practices to work from.
What "AI-ready" data actually means: a fast overview of why most data stacks were built for human analysts, not AI agents asking questions around the clock, and where that gap shows up in practice
Context engineering in practice: how Fivetran + dbt establish a governed context layer so AI tools retrieve the right information instead of guessing, with real examples of what works and what doesn't
Integrating AI tools with trustworthy data: leveraging the dbt MCP server and agent skills so that a client's AI tools return accurate answers without burning excess tokens or compute
What this means for your business: how to turn this readiness work into an actual services conversation with clients who are already experimenting with AI on their own data, before they realize the foundation underneath it isn't ready
Justin Erickson
Kai Lee
Justin Erickson
Kai Lee
Ever watched an AI tool confidently hand back a wrong answer because it was pulling from stale, disconnected data? That's not an AI problem; it's a data foundation problem. In this session, we get into the specifics of what actually makes data trustworthy for AI, with real examples of how to structure it and connect it with Fivetran + dbt so AI tools stop guessing.
Who should tune in:
- Solution architects and delivery consultants: if you're the one building or delivering AI use cases and designing client data architectures today, this is where the ground is shifting under AI initiatives. Come see what "AI-ready" actually requires before someone else figures it out first.
- Practice leads and account owners: your clients are already experimenting with AI on their own data. This is the technical grounding that turns that curiosity into a services conversation you can actually sell, and it gives your delivery teams a shared foundation of best practices to work from.