Autonomous in the loop. Accountable at the gates. An agentic offering by Plainsight

dbt, or your own framework. You choose; we build it.

Already deep into dbt? We build dbt. Built your own framework over the years? We build that.

Why we offer both

Standards that survive the migration.

A migration shouldn't force you to abandon the way your team works, or stop you adopting a cleaner one. So the fleet documents your estate once into the knowledge base, and the builders deliver it in whichever flavor fits: a fresh, standardized dbt project, or your own framework, faithfully reproduced. Same migration, same gates, same loop, your standards either way.

dbt, or your own framework
dbt
testsdocslineage
your framework
namingorchestrationcontinuity
We build both. On Microsoft Fabric or Databricks.
Two flavors, one decision

When to choose which.

dbt logo

Build in dbt

For a fresh, standardized build. The builders generate a full dbt project (your layers, macros, snapshots, tests and docs) on Microsoft Fabric or Databricks. dbt itself is open-source and free to run, so the flavor adds no licence cost.

  • A completely new implementation or a clean re-platform
  • You're standardizing your whole team on dbt
  • You want to discard tangled legacy orchestration and rebuild the intent cleanly
  • You want tests, docs and lineage generated out of the box

Build in your framework

For teams that built their own way of working. The builders generate into your existing patterns, naming, error handling and orchestration: it's yours, not a Plainsight product.

  • You've built your own way of working over the years and it works
  • Your naming, error handling and orchestration are standards you trust
  • Your teams are trained on your existing patterns
  • You want continuity and minimal change-management
Let your estate decide

Which way does your estate lean?

Leanings, not laws, and you don't have to guess up front.

dbt your framework
Greenfield or a clean re-platform Leans dbt

you're not preserving legacy orchestration, so a modern, tested transformation framework is the simplest place to land.

Throwing away tangled legacy logic Leans dbt

rebuild the documented intent as a clean model DAG instead of porting the mess.

Want to avoid added tooling licence cost Leans dbt

dbt Core is open-source and free to run, so the build adds no software licence on top of your platform.

Years invested in your own framework Leans your framework

your conventions are an asset to keep, not a cost to rebuild.

Teams trained on your existing patterns Leans your framework

lower change-management, people keep working the way they already know.

Mixed estate, or genuinely unsure A bake-off decides

the fleet documents once and can build a representative slice both ways before you commit.

Still unsure? The fleet documents once, then builds a representative slice both ways: a bake-off, with real numbers, at no risk to the rest of the estate.

Both flavors, both platforms

dbt and your framework run on Fabric and Databricks.

The flavor is never in the URL, it's a choice at build time. See exactly what each produces on each platform:

Microsoft Fabric logo

Microsoft Fabric

dbtyour framework

What the builders generate on Microsoft Fabric, in either flavor.

Migrate to Fabric →
Databricks logo

Databricks

dbtyour framework

What the builders generate on Databricks, in either flavor.

Migrate to Databricks →

Or jump straight to a source-and-target pair:

Good questions

dbt or your framework, answered.

Can we mix the two flavors?

Most teams pick one, but the flavor is decided at the design gate and can differ where it genuinely makes sense, the knowledge base underneath is the same either way.

Can we switch flavor later?

Cheaply, before the build starts. The knowledge base and the Target Design Spec carry over unchanged; only the builder path differs. Switching after a build means regenerating those assets.

Does dbt run on both Microsoft Fabric and Databricks?

Yes. On Fabric the builders generate a dbt project targeting the Warehouse; on Databricks a dbt project targeting Databricks SQL, with incremental models using MERGE and snapshots for slowly changing dimensions.

Will the your-framework build really match our conventions?

That is the whole point. The Architect encodes your naming, error handling and orchestration into the Target Design Spec, and the builders generate into your existing patterns, so the output looks like your team wrote it.

Does the dbt flavor add any licensing cost?

No. The builders generate a standard, open-source dbt (dbt Core) project that you own and run yourself, so there's no licence fee for the framework itself, you pay only for the Fabric or Databricks platform underneath.

Do you tie us to a particular dbt edition?

No. “dbt” here means a standard, open-source dbt project (dbt Core): your layers, macros, tests and docs. It's free to run, and we don't lock you into a paid edition or vendor.

Let's talk

dbt, your framework, or a bake-off?

Tell us how your team works today and we'll recommend a flavor, or build a representative slice both ways so the numbers decide.

Plan my migration

A short form, no spam. We usually reply within one business day.

Plan my migration