FOR DATA & PLATFORM ENGINEERS

Auto-generated.
Yours to edit.

DataAgents writes the transformations, the orchestration, and the deployment manifests. You open them in a visual studio, change what the platform got wrong, and ship without rebuilding the dbt project.

Open the visual studio →Read the architecture

Visual SQL studioSync & orchestration monitoringML deployment consoleGit-backed publish flow

contribution_margin.sql

churn_risk.py

orders.yml

METRICS

▾contribution_margin.sql

·arr_governed.sql

·cac_blended.sql

PIPELINES

·stripe.fees.sync.yml

·shopify.orders.sync.yml

MODELS

·churn_risk.py

·ltv_forecast.py

1-- generated · review pending · owners: finance, growth

2CREATE METRIC contribution_margin AS

3 SELECT

4 date_trunc('day', o.created_at) AS grain,

5 sum(o.net_revenue - o.refunds - f.fees - o.cogs)

6 / nullif(sum(o.net_revenue), 0) AS margin

7 FROM shopify.orders o

8 LEFT JOIN stripe.fees f USING (order_id)

9 GROUP BY 1

10-- you edited this on Apr 24 · diff vs proposed · 4 lines

build · okdiff +4 / −2publish to · main

31,402

SYNCS / 24H

Successful, retried, and failed windows visible in one orchestration view.

2m 14s

FRESHNESS P95

Warehouse-style monitoring without warehouse-style babysitting.

LIVE MODELS

Production, canary, and rollback state tracked like services, not notebooks.

VISUAL STUDIO

The transformations the platform writes.
Editable, with diff and rollback.

Every auto-generated metric and pipeline lands in a Git-backed studio. Inspect the SQL, change what is wrong, hit publish. The change ships through a two-reviewer flow with full lineage on the impact.

A · PROPOSE

Auto-generated

The platform writes a metric, transformation, or pipeline based on the source schema and your team's vocabulary.

→

B · EDIT

You change what is wrong

Visual SQL editor with inline diff, lineage on hover, and the SQL plan attached to every metric.

→

C · PUBLISH

Review, ship, rollback

Two-reviewer publish, Git-backed history, and one-click rollback if a metric breaks downstream consumers.

For platform & data engineering teams

You did not become an engineer
to write reverse-ETL pipes.

Open the visual studio, ship the metric, monitor the orchestration, deploy the model in one product, with a real Git history behind it.

Open the studio →Read the architecture

Auto-generated.Yours to edit.

The transformations the platform writes.Editable, with diff and rollback.