Model your data in Weld, GitHub or dbt

Write SQL models in Weld, sync them from GitHub, or keep your dbt project and let Weld trigger it when fresh data lands. Build them in Claude Code, Cursor or VS Code, and run them in your warehouse.

Start for free Read the docs

Models live in Git
Runs in your warehouse
SOC 2 Type II

Click anything to explore

Initial sync in progress...

Initial syncs take time as they process all historical data. Querying is unavailable until syncing is complete.

demo_customeraov_daily

github_models aov_daily Published

Sourced from acme/analytics · models · main, read-only in Weld Transform.

1select

2 date_trunc('day', order_ts) as day,

3 sum(revenue) / count(*) as avg_order_value,

4 count(*) as orders,

5 sum(revenue) as revenue

6from {{github_models.demo_order_line}}

7group by 1

Claude Code

What was average order value yesterday?

●weld·list_transforms⎿8 models from GitHub
●weld·get_transform⎿github_models.aov_daily
●weld·run_query⎿30 rows · 0.8s

✓94.47 on 412 orders, from aov_daily.

>Ask Weld to move or model your data

ConnectedWeld MCPGitHub

Trusted by data and AI teams

Features

Keep the workflow you have. Weld runs it on fresh data.

Bring your editor, your dbt project and your orchestrator. Weld handles ingestion and can start your models the moment a sync finishes.

Built in your AI editor

Write models with Claude Code, Cursor or GitHub Copilot, push to GitHub and they deploy to Weld automatically.

git pushmain · models/revenue_by_channel.sql

dbt Cloud or dbt Core

Weld triggers your dbt Cloud jobs through webhooks when a sync finishes. On dbt Core, your own orchestrator receives the webhook and runs dbt.

webhookdbt Cloud job triggered after sync

Weld SQL workspace

A built-in SQL editor with GitHub sync, lineage, tests, scheduling and AI assistance, for models you would rather keep in Weld.

lineageorders → revenue_by_channel → hubspot

Transforms API

Create, update and publish models over REST, read their version history and run them from any CI/CD pipeline.

POST/transforms/{id}/publish

From commit to rows

Push a model. Weld runs it on fresh data.

The path a model takes from your editor to the warehouse, one step at a time.

Pushed from Claude Code

branch: main
commit: add revenue_by_channel
file: models/revenue_by_channel.sql

Transforms API endpoints, for CI/CD and agents

sources feeding your models

syncs every day that can trigger them

Tools

Write where you work. Run where your data lives.

Models live in Git and run on the warehouse you already pay for. Weld starts them when fresh data lands.

Where you write

Claude Code
Cursor
Copilot
GitHub
dbt

Data TransformationsRuns when a sync finishes

Your warehouse

BigQuery
Snowflake
Databricks
Redshift
ClickHouse
MotherDuck

Set up dbt with Weld

Customer story

“We’re still a team of three, but we’re often doing far more than the equivalent of three full-time employees. That’s down to how we're able to leverage systems, data, and processes.”

Michael Howes

Head of Data & Insights at Dishoom

How Dishoom scaled data operations without scaling its team

Read the story

The platform