Pipelines

Pipelines that build, test and explain themselves.

Transformations, orchestration and backfills in one place. Models are Python and SQL in your repo; the platform schedules them, runs them on managed compute and shows the lineage of every table.

Start a pilot Join the waitlist

Pipelines as code

Assets, dependencies and partitions are Python and SQL in your repo. The platform builds the graph, schedules it and runs it on managed compute.

Backfills and replay

Backfill any range of days from the console, CLI or agent. Runs are isolated, logged and replayable.

Lineage and run history

See what feeds what, which run produced a table and what failed upstream, in the console and the CLI.

Preview before merge

Every pull request gets a preview of the change before it reaches production data.

In code

pipeline/models/conversion_rates.py

@s.model(
    tier=s.Tier.Silver,
    partitioning=s.Partitioning.Unpartitioned,
    inputs={"leads": "silver/lead_snapshots"},
    schema={"segment": "!string", "requests": "!int64", "sqo_rate": "!float64"},
)
def conversion_rates(leads: ir.Table) -> ir.Table:
    """SQO rate of last year's demo requests by segment."""
    return leads.group_by("segment").aggregate(
        requests=_.count(),
        sqo_rate=_.is_sqo.mean(),
    )

Pipeline SDK reference →

Use it, install it, or build on it

Every block on the platform comes three ways. Most teams use all three.

Use it

Run, backfill and inspect jobs from the console or CLI.

Install it

Install a starter or an app that ships its own tested pipeline.

Build on it

Your agent writes models and jobs; you review the pull request and its preview.

Replaces dbt Cloud, Airflow and Astronomer.

Install from the marketplace

dbt

Your existing dbt project, models and tests.

Connector · Connect or replace

Analytics workspace

A repo with pipeline, connections and a deployed app.

Starter

Data Quality Monitor

Catch broken sources and tracking before anyone reports on them.

Extension

Browse the marketplace →

Solutions built with it

Marketing analytics

Attribution and budget optimization on first-party data

Lead scoring

Grade leads A to D against standards you can read

Warehouse migration

Move off Snowflake or Databricks with proven parity

Enterprise pilot

Start a pilot.

Want pipelines set up for you? Our forward-deployed engineers build it for you in your workspace, as code you own.

One solution, on your data, in your own workspace
First version live in two weeks
Parity proven before anything is switched off
Fixed scope and price, credited to your contract if you continue

Not ready for a pilot? Send the form anyway with your question. An engineer will reply, not a sales rep.

How the pilot works

Your company’s context, ready for every agent.

$ segmentstream workspace create acme --region eu

▸ workspace ready · data, objects, metrics and access rules in one place

▸ connect Claude, ChatGPT or Cursor with one connector

▸ tell your agent: “add our billing API as a source”

Start a pilot

Our engineers build the first version on your data in two weeks, at a fixed price.

Or join the waitlist for self-serve:

Pilots are open now. Self-serve opens to the waitlist this fall.