Skip to content
Molecular AI
Book a scoping call

Engagement · Context Layer

Context Layer

“Make the warehouse answer questions in plain English, against definitions we all agreed on.”

An entity and metric layer over the warehouse you have — what a customer, an order or revenue actually means — so a plain-English question resolves against a definition your team approved rather than against whichever raw table looked right.

Duration
Scoped on the call
Starts with
A Warehouse Audit
Engines
Snowflake · Athena/Glue
Access
Read-only, then a staging schema

In scope:

  • An entity model: what a customer, an order or a shop means across your tables, and where those meanings currently disagree
  • One agreed definition per metric, with the SQL behind it and the lineage under it
  • Every conflicting definition surfaced before it is resolved, so the choice is made deliberately rather than by whoever queried first
  • The definitions written back as models your team owns, generated and run against your warehouse
  • Plain-English querying and charting over the result — a request for a measure returns SQL grounded in these definitions
  • A working session to settle the definitions, with the people who disagree in the room

What we need

  • Read-only warehouse credentials
  • Your dbt project, a folder of SQL, or a Glue catalog
  • Named owners who can settle a definition when two teams disagree — this is the real prerequisite, not the credentials

How long

Scoped on the call. Driven by how many metrics are in scope and how many of them are contested.

Out of scope:

  • Ruling on which definition wins. We surface the conflict and recommend; your owners decide
  • Replacing your BI tool or re-pointing existing dashboards
  • Deploying anything into production

What we mean when we say tested.

We generate the models, run them against your Snowflake or Athena connection, and check that each one returns rows, that its primary key is unique, that no column comes back completely empty, and that the types are what we said they'd be. On top of that we run dbt's own generic tests — not-null, unique, accepted values, and foreign keys against their parent — and hand you the schema.yml for them, so the tests we ran are tests you keep. Where a test can't be evaluated soundly it comes back marked skipped, never passed. We don't run a full dbt build, and we don't compare your old numbers against the new ones row by row. If a figure has to reconcile exactly, that's a conversation for the call.

Not here yet

Built, not yet switched on The plain-English querying and charting surfaces are built, but not yet switched on for beta accounts. Until they are, this engagement is delivered by us against your warehouse rather than something your team self-serves.

Written in future tense on purpose. Anything on this list is something we intend to ship, not something you would be buying today.

Scope the engagement in thirty minutes.

Bring the domain that causes the most difficulty. If it turns out this is not the right engagement for you, we will say so and point at the one that is.

See the other engagements · What Molecular AI is