Manual pipeline

Dataflow-style

Source → Filter → Map → Aggregate → Sink. No API key needed.

Source

—

Filter

Optional

Map

Optional

Aggregate

Optional

Sink

New table

Source table

Filter column / op / value

Keep columns (blank = all)

Aggregate

Group by

Sink table name

Pipeline name

Run log

Copy job

Full / Incremental / Stream

1. Source

2. Data

3. Destination

4. Settings

5. Review

Copy job mode

Choose how data is copied. Applied every run. After creation you can schedule and monitor.

Full copy
Copy all data at once Incremental copy
Initially all data, then only changes Stream copy
Copy data streams from real-time events

Incremental column

Source Table	Incremental column

Copy log

Glen Bot

optional

Requires Groq key in Ops → Keys. Manual & Copy never need this.

Source table

Directive

Publish returned table to Lakehouse

Saved pipelines

Clusters

Compute pools

Create local compute clusters to group connections, pipelines, and lakehouse work — Fabric-style capacity pools (simulated in-browser).

Cluster details

Name

Type

Size

Auto-stop (minutes idle)

Description

Attach connections

Connections

Supabase · Mongo · AWS · GCP · Azure · MySQL · HTTP

Connect external databases and APIs. Credentials stay in this browser only. Ingest into Lakehouse tables.

Connection details

Name

Type

Multi-database workspace

—

No table open

Open tables from the left — multiple tabs supported

Open several connections and tables at once. Each table opens in its own tab.

Ingest into Lakehouse

Source table / collection

Target lakehouse table

Row limit

Pipeline schedules

Clock-style (runs while this tab is open).

Saved pipeline

Every N minutes

Manual pipeline

Copy job

Glen Bot

Saved pipelines

Clusters

Connections

Multi-database workspace

Pipeline schedules

Confirm