Agentic AI has opened up the possibility to build at lightning speed, while giving us agents that can handle uncertainty along the way.  In this presentation, we'll explore what that looks like specifically for data platform migrations. We'll migrate a legacy DWH to a new modern platform live on stage, describing the process behind it. Our approach here consists of two steps: first, the readiness step creates an inventory of what is there and highlights difficulties ahead of the migration; next, the migration step takes that inventory and uses specialized agents to convert the code, delivering 100% schema and logic migration up to 60% faster than manual, traditional approaches. Alongside the live demo, we'll draw on examples from real customer projects to show how this plays out in practice outside a stage setting.

Who this is for:
Data leaders, architects, and engineers responsible for legacy platform migrations and modernization. Particularly those working with fragmented pipelines, unclear data quality, or governance that was bolted on after the fact rather than built in. Also relevant for teams evaluating how agentic AI fits into data engineering work beyond isolated coding assistants.

What they'll walk away with:
•    A concrete look at what an agentic migration actually involves in practice, demonstrated live rather than described in slides
•    A view of how a readiness assessment can surface migration risks before code conversion begins, rather than mid-project
•    A clearer picture of where AI agents can operate with minimal oversight versus where human judgment remains essential 
•    A reference point for scoping their own migration