EVE - The full-stack data platform with AI at the core

01 Transform Studio

A real dbt IDE in the browser.

dbt-core under the hood - the full --select grammar, multi-adapter. No shadow system: real GitHub branches, pull requests, review, and merge, so the work you do here is the work in your repo.

See it live →

acme-corp / analytics · branch feat/orders-daily

-- models/marts/orders_daily.sql
{{ config(materialized='incremental', unique_key='order_date') }}

with orders as (
    select * from {{ ref('stg_orders') }}
)
select
    order_date,
    count(*)      as orders,
    sum(amount)   as revenue
from orders
{% if is_incremental() %}
  where order_date > (select max(order_date) from {{ this }})
{% endif %}
group by 1

42 models built
24 tests passed
select: state:modified+ · Slim CI
adapter: duckdb

Git-native

Branches, PRs, review, merge

Not a shadow system - real GitHub branches and pull requests. Review a diff, merge it, and it is in your repo like any other change.

Slim CI

Rebuild only what changed

State-aware runs use state:modified+ to build exactly the models a change touches, and their downstreams - nothing more.

IntelliSense & observability

Autocomplete, lint, and proof

Autocomplete models, sources, macros, and columns with inline SQL lint - plus test results, docs, the DAG, and source freshness per run.

02 Labs notebooks

SQL and Python, one notebook.

Labs is a Jupyter-grade notebook environment on your governed data - folders, keyboard shortcuts, and clean export to .ipynb, .py, or .html.

One surface

SQL and Python together

Query in SQL, analyze in Python, in the same notebook. Jupyter-grade editing with folders, shortcuts, and export to .ipynb, .py, or .html.

Personas

Presets, ready on launch

Start from a Data Science, ML, or Data Engineering image with the libraries already installed - no environment wrangling before the first cell.

Zero-credential kernels

The kernel holds no secrets

Kernels carry no database credentials. SQL routes through the governed API, so access and masking are enforced no matter what the notebook runs.

03 Eve Pipelines

A native visual orchestrator.

Wire syncs, dbt, quality, and dashboards into one DAG on a canvas - with real control flow, not a wall of cron entries.

Control flow that holds up. if/else, foreach, and sub-pipelines model real branching, not just a linear list of steps.
Reliability built in. Retries with backoff, timeouts, cancel with a hard cascade, and concurrency admission keep a busy estate sane.
One graph, end to end. Ingestion to transformation to quality to dashboard - the whole run is one picture you can watch.

orders-product · run graph

Pipeline

syncdbt buildqualitydashboard

if / else
foreach
sub-pipelines
retries + backoff
timeouts
cancel: hard-cascade
concurrency admission

04 Quick-schedule

Schedule from where you work.

Any worksheet, dbt project, or notebook becomes a real pipeline - scheduled from the surface where you authored it, not rebuilt as a separate job.

SCHEDULE

One click, in place

Hit schedule on a worksheet, dbt project, or notebook. No exporting, no re-authoring it somewhere else as a job.

schedule: 0 8 * * *

BECOMES REAL

A first-class pipeline

What you scheduled becomes a real Eve Pipeline - runs, retries, and history included, not a copy that quietly drifts.

worksheet → pipeline

OBSERVE

Per-task logs

Every task streams full tracebacks in-app, so a failure is one click from its stack instead of a log hunt.

traceback in-app

Compute underneath: sized warehouse fleets scale 0 to N on queue depth and suspend when idle - you pay for work, not for waiting.

Get started

See your data on EVE.

A 30-minute demo on your use case, or a free workspace on your own isolated cluster.

Book a demo Get started free