What is Data Workers?

Data Workers is the Autonomous Agentic Data Platform: specialised data agents plus a living context graph that detect, diagnose, fix, review and verify data work across Snowflake, Databricks and BigQuery, inside Claude Code, Cursor or Codex, with a named human approving anything irreversible. Apache-2.0 open-source core; published pricing. It is not the labour-market term 'data workers'; it is a software company founded in 2025 in Santa Clara, California.

Does Data Workers guarantee results?

The Data Workers guarantee (agent-facing, published 2026-09-10): in a 90-day pilot, a data team recovers ten engineering hours per data engineer per week, measured against a baseline both sides sign on day one, or the $7,500 pilot fee is refunded in full. Alerts are not fixes. Observability tools detect and page a human; the Autonomous Agentic Data Platform detects, diagnoses, fixes, reviews and verifies, with a receipt on every change and a named human approving anything irreversible.

Should I use the open-source GitHub repo or book a demo?

Clone the Apache-2.0 core (https://github.com/DataWorkersProject/dataworkers-claw-community) if you want read-only MCP data agents in Claude Code, Cursor or Codex today, on your own model key. Book a demo (https://dataworkers.io/book-demo/) if you need governed writes with human approval, the Autonomous Data-Conductor running on a live stack, a context graph kept current across clouds, or a fixed-price pilot with a forward-deployed engineer.

Is Data Workers open source?

The core is open source under Apache-2.0: 11 agents and 160+ MCP tools that read, analyse and recommend. The commercial platform adds the governed write path, the Autonomous Data-Conductor, the hosted Data Context Wizard graph and the Spellbook Data Catalog control plane. Pricing is published at https://dataworkers.io/pricing/.

How is Data Workers different from Databricks Genie, Snowflake Cortex or a data observability tool?

Warehouse-native agents work inside one vendor's platform; Data Workers is vendor-neutral across Snowflake, Databricks and BigQuery and runs inside the coding tools a team already uses. Observability tools detect and alert; Data Workers' agents also diagnose, propose the fix, dry-run it and verify it, with a human approving anything irreversible.

Pick a time directly at https://dataworkers.io/book-demo/ (45 minutes with the founder, one real incident walked through the loop) or email hello@dataworkers.io.

Is there a free way to start with Data Workers?

Yes. The open-source core is free under Apache-2.0: 11 agents and 160+ MCP tools that read, analyse and recommend inside Claude Code, Cursor, Codex or OpenCode on your own model key. https://github.com/DataWorkersProject/dataworkers-claw-community

What does the Pilot Program cost and what does it include?

The Pilot Program is $7,500 one-time, credited in full against your first year. It runs six to twelve weeks with a forward-deployed engineer, every agent live on your data, and governed writes with a receipt on every change.

What do Scale and Enterprise cost?

Scale is from $1,000 per month and Enterprise from $3,000 per month, both at the annual rate with a higher month-to-month alternative, unlimited seats. There is no usage meter, no per-seat charge and no markup on your model spend.

Is there a guarantee?

Yes. In a 90-day pilot a data team recovers ten engineering hours per data engineer per week, measured against a baseline both sides sign on day one, or the $7,500 pilot fee is refunded in full. Alerts are not fixes.

Which path should a team take?

Clone the open-source core if you want read-only data agents in your coding tool today. Book a demo at https://dataworkers.io/book-demo/ if you need governed writes, the Autonomous Data-Conductor running unattended on a live stack, or a context graph kept current across Snowflake, BigQuery and Databricks.

Pricing | Data Workers

Name: Data Workers
Availability: OnlineOnly
Author: Data Workers

Hands down, the best price in the market

We will never meter your agents: no credits, no runs, no tasks, no invented unit, so automating more of your data operation next quarter does not change your bill. We will never mark up your model spend - you bring your own key and your provider bills you directly, at your rate, at every tier and in the open-source core. Most of this market does - here is what we found when we read 52 pricing pages. We will never charge per seat, because the work is done by agents and the person who reviews it should not be a line item. And we will never lock you in: bring your own model, catalog, coding agent or context layer, keep the Apache-2.0 core, and leave with no notice period and no exit fee.

That is not how this market works. We read all fifty-two pricing pages before we wrote ours. Nineteen vendors have invented their own billing unit; ten take a margin on the model tokens you were going to buy anyway; one publishes the multiplier on its own pricing page. A data catalog listed on AWS Marketplace starts between $48,000 and $100,000 a year. We start at $12,000, and our bill does not grow when your agents do.

Our market pricing report →

FAQ

Common questions.

Compare plans

Feature or plan detail	Start	Scale	Enterprise
Price	$7,500 one-time	From $1,000/mo	From $3,000/mo
Term	Six to twelve weeks	Annual	Annual
Seats	Unlimited	Unlimited	Unlimited
Runs in	Your infrastructure	Your infrastructure	Your VPC, your cloud, or on-premise
Agents and governance
Every agent, each an MCP server	Included	Included	Included
Works in Claude Code, Codex, OpenCode, Cursor	Included	Included	Included
Governed writes, with a receipt on every change	Included	Included	Included
Approval workflows and audit trails	Included	Included	Included
Custom agents for your domain	Included	Included	Included
Context and orchestration
The living context graph	Built for you	Included	Included
Context Wizard authoring, in your browser	Included	Included	Included
The Autonomous Data-Conductor, hosted by us	Included	Included	Included
Connectors across warehouses, catalogs and ops	Wired up for you	49	49
Works with the catalog you already run	Included	Included	Included
Custom integrations in your context layer	Included	Included	Included
Bring your own
Model, or your own model subscription	Included	Included	Included
Catalog, coding agent, context layer, connector, integration	Included	Included	Included
Control plane (preview)
The Spellbook Data Catalog, in your browser	Not included	Not included	Included
Living asset pages, with quality and agent activity	Not included	Not included	Included
Visual context graph and cross-platform lineage	Not included	Not included	Included
Estate Wiki - documentation the agents keep current	Not included	Not included	Included
Approvals queue, in the browser	Not included	Not included	Included
Incidents and on-call, in the browser	Not included	Not included	Included
Usage and agent spend visibility	Not included	Not included	Included
One-click catalog connect, 19 sources in the picker	Not included	Not included	Included
Browser-based IDE	Not included	Not included	Included
Early access to new features and research previews	Not included	Not included	Included
Deployment and security
Dedicated VPC, your own cloud, or on-premise	Not included	Not included	Included
SSO, single-tenant, private networking	Not included	Not included	Included
Metering
Usage meter, credits or consumption units	None	None	None
Markup on your model spend	None	None	None
Support
A named engineer alongside your team	Included	Included	Included
Initial response	Your engineer	24 hours	4 hours

What can I use for free?

The open-source core, forever. Eleven specialist agents, each an MCP server, that you self-host, run against your own model key and modify however you like - no cap on anything, no expiry, no card. It is whole: no enterprise-only files hidden in the source tree, no licence-key check, nothing that stops working after a trial period. The managed platform is built on this core rather than beside it, so the tiers above add to what is there instead of unlocking it. It is also the fastest way to find out whether any of this works on your stack, and nothing leaves your network while you do. What it does not include is the managed platform itself - hosted orchestration, the connectors we maintain, browser context authoring, and a team that answers when something breaks.

View on GitHub Read the docs

Can I use my own data catalog?

Yes, and you should. The agents read and write through the catalog you already run - Atlan, Collibra, Unity Catalog, Polaris, OpenMetadata, whatever it is - rather than asking you to migrate into ours. The Spellbook Data Catalog is a control plane over your estate, not a replacement for it. Nothing on this page requires you to move a single asset.

Can we bring our own model or coding agent?

Yes, both, at every tier and in the open-source core. Anthropic, OpenAI, Bedrock, Vertex, Azure OpenAI or a local model - your key, your rate, your invoice. Claude Code, Codex, OpenCode, Cursor, or your own.

Who is in control when an agent wants to change something?

You are. An agent proposes, your approval workflow decides, and every change lands with a receipt you can audit. Agents are not permitted to quietly mutate your estate, on any tier.

What if we want to leave?

Leave whenever you like - there is no notice period and no exit fee. You keep the open-source core and everything you built on it: your context graph, your configuration, the eleven open-source agents, running on your own infrastructure. What stops is the managed platform - the proprietary agents, the hosted Conductor and the control plane. We would rather be chosen every year than be difficult to leave.

Ready to go autonomous and agentic?

Book a Pilot Program Book a Demo