Platform
One plane, from the source to the proof.
Most data stacks are five vendors stapled together, each with its own copy of the data and its own idea of what happened. This is the same five jobs done by one platform, on your infrastructure, against one catalog.
- Runs
- Your infrastructure
- Core license
- Apache-2.0
- Outbound calls
- None required
- Telemetry
- None
What connects to what
Move, govern and answer are a path. Operate and prove are bands that run underneath all of it — which is exactly why bolting them on from a separate vendor never quite works.
Revoluzen data plane
your network · no outbound dependency
Every box runs on infrastructure you control. The dots are your data; none of them leave the diagram.
Get the data in, on a schedule you trust
Visual and code pipelines across 100+ connector types, with retries, idempotent backfills and per-step resume.
What it covers
- 01100+ connector types, from Postgres and S3 to SaaS REST APIs
- 02Visual DAG canvas with 40+ step types, plus SQL and Python steps
- 03Change data capture via Debezium, for live replication
- 04Enterprise sources: SAP, SAP HANA, Oracle, Teradata, DB2, Informix
Revoluzen Flow
Pipeline orchestrator
Build data pipelines on a visual canvas or in code, run them on your own machine, and see exactly what every step did.
Revoluzen Flow Enterprise
Enterprise edition
Everything in Flow, plus the parts an enterprise platform team asks for on day one: CDC, enterprise sources, workspaces and release control.
Model it once, govern it everywhere
A medallion lakehouse with dimensional models, lineage and a governed SQL endpoint — not a spreadsheet of table names.
What it covers
- 01Bronze / silver / gold medallion layers with enforced contracts
- 02Dimensional modelling and conformed dimensions as first-class objects
Answers that cite the model they came from
Dashboards, reports and a governed query layer where every number traces back to a defined metric.
What it covers
- 01Dashboards and scheduled, paginated reports from one semantic model
- 02Governed answers: every figure traces to a defined metric and query
Run a fleet, not a pile of installs
One control plane for every environment: promotion between dev and prod, approvals, run history and alerting.
What it covers
- 01Fleet inventory across environments, versions and health
- 02Promotion between development and production with a review gate
- 03Live pipeline runs and the DAG behind them, from Revoluzen Flow
- 04Approvals with the change in front of you: diff, versions, multi-step sign-off
Revoluzen Console
Control plane
One place to see and operate every Revoluzen instance you run — across laptops, servers, sites and customers.
Revoluzen Mobile
Companion app
The platform in your pocket: watch pipeline runs, get the alert that matters, and approve a release from your phone — through your own network, never ours.
Prove it ran, months later
Signed execution receipts, so an auditor can verify what ran, on what input, with what result — without trusting us.
What it covers
- 01Ed25519-signed verifiable execution records
- 02Verification needs the receipt and a public key — not our servers
Your model. Your keys. Your network.
We build AI features into the products because they genuinely help — drafting a pipeline, explaining a failure, answering a question about your own instance. We do not build a dependency on a specific provider, and we do not route your prompts through us.
Fully offline is a supported mode
Point the Copilot at a local model and nothing leaves the machine. This is a first-class path, not a degraded fallback — the product works without a frontier model, it simply drafts less fluently.
Or your own provider, your own account
Use the hosted provider you already have a data-processing agreement with, under your own keys. We do not resell tokens, meter your usage, or sit in the middle of the call.
Scoped and permission-aware
All 26 Copilot tools read only your own pipelines, runs and catalog, and respect the permissions of whoever is asking. The assistant cannot see more than the person using it.
The choice of model and provider belongs to you. It is your data, your compliance posture and your commercial relationship — not a lever for us to pull.
The same platform at three sizes
Nothing structural changes between an evaluation on a laptop and a two-site production deployment. You are not rebuilding when you grow — you are changing where it runs and what backs the metastore.
One laptop
A pip install and a browser tab. SQLite storage, loopback binding, no account and no outbound calls. Build a real pipeline against a real source in an afternoon.
- Storage
- SQLite file
- Bind
- 127.0.0.1:8001
- Users
- One
One server
An internal host with a Postgres metastore and your identity provider in front. Scheduled production pipelines feeding the lakehouse, with real users and real permissions.
- Storage
- Postgres
- Identity
- Your IdP / SSO
- Users
- A team
A fleet
Multiple environments or sites under one control plane, with promotion gates between dev and prod, central run history, and high-availability topologies where you need them.
- Storage
- Postgres, HA
- Control
- Revoluzen Console
- Users
- Many teams
The honest way to evaluate this is to run it.
Revoluzen Flow is Apache-2.0 and installs in one command. Build something real with it first; the commercial conversation is easier once you know whether the engineering suits you.