Notes from the VeLens team

Product updates, Marketing Cloud engineering notes, and the thinking behind building an AI-native layer for SFMC.

What we're writing next

No filler posts, just the topics we actually plan to publish. Each one lands here as it is written.

Why we built an AI agent inside Marketing Cloud

The case for an assistant that lives on top of SFMC instead of in yet another dashboard, and what "in the flow" really means for daily campaign work.

How VeLens keeps subscriber data away from the LLM

A look at the tool boundary that masks row-level data, why the model only ever sees schemas and counts, and how writes stay confirm-gated end to end.

SFMC SQL patterns the AI copilot learned

Schema-aware SQL in Query Studio, from ENT. shared DEs to system data views. The patterns that make generated queries actually run.