New

Turn a hand-built transformation layer into a versioned semantic model

Point the dltHub AI harness at the SQL you already run and get an ontology, a canonical data model, and Chat-BI that reasons like an analyst.

Put this blueprint to work

Talk it through with the dltHub team and we will help you ship it to production.

Your coding agent

Claude Code, Cursor or Codex

operates through

the agentic layer

dltHub AI harness

Agentic primitives to build, run and fix pipelines

dltHub context catalog

Lineage, schema, data quality, governance, run state

grounds and runs

What you already run

HubSpot
Op
Operational database
Ha
Hand-built SQL transformations

the managed infrastructure layer

The AI harness ontology toolkit reads your SQL as a spec, drafts an ontology, and generates the canonical model and declarative transformation layer. Execution moves to dltHub.

Semantic model & Chat-BI

One versioned ontology behind your dashboards, exports, and Chat-BI. Every concept defined once, the same way everywhere.

Your HubSpot, operational database, and hand-written SQL in. A versioned ontology, a canonical data model, and Chat-BI out.

How it works

Most first-gen data stacks follow the same pattern: a CRM like HubSpot, an operational database, and a transformation layer someone built by hand and now has to keep patching. This Blueprint turns that into a versioned semantic model that anyone on your team can configure, avoiding a multi-month rebuild or new hires.

The dltHub AI Harness provides an ontology toolkit you can point at your existing pipeline. It drafts an ontology of a handful of clean concepts (Person, Account, Interaction, Deal, Product Event) from your messy or unorganized data, then generates both a Canonical Data Model and a declarative transformation layer from the ontology. Execution shifts to dltHub's managed runtime, so the legacy orchestration and the schema enforcement it demanded are retired.

From there, Chat-BI runs against the same semantic model feeding your dashboards and exports. Since the ontology actually declares meaning such as entities, relationships, metrics, or business rules, Chat-BI can reason about your business like an analyst would. Every concept now has one consistent definition.

Key features

Reads your existing SQL as a spec and reverse-engineers a draft ontology from it
Consolidates scattered tables and fields into a small set of canonical concepts, then generates a Canonical Data Model and declarative transformation layer directly from them
Runs execution on dltHub's managed runtime, an agentic layer on top of the warehouse you already have, with the legacy orchestration retired
Gives every metric and entity one versioned definition, so agents, analysts, and auditors read the same artifact
Serves Chat-BI from that same semantic model behind your dashboards, so your team gets real answers instead of text-to-SQL guesswork

How to get it

The ontology toolkit ships with the dltHub AI harness.
Contact dltHub for a 30-minute scoping call on your existing transformation layer.

Pricing

dltHub from $1,190/month: the platform the pipelines and transformations run on.
Custom: we build the ontology and semantic model together with our customers. Set up a 30-minute scoping call if you are interested.

Built by

dltHub
NAVITTechnology partner
Martin Miodownik · CTO & Co-Founder at NAVIT

Put this blueprint to work

Talk it through with the dltHub team and we will help you ship it to production.

Find more blueprints