SourceX
Proprietary enterprise data for AI training and evaluation
A catalog of the business data SourceX sources for AI labs: support tickets, private code and reviews, call recordings, CRM outcomes, contracts and workflow records, sourced to order from the companies that created them. Each listing explains what the data contains, where it comes from, how it’s prepared and how to request it.
Categories
| Category | What it covers |
|---|---|
| Customer support data | Customer support datasets pair real customer problems with the steps agents took to resolve them, often with satisfaction and QA scores attached. |
| Call and meeting recordings | Recorded calls and meetings capture how people actually talk at work, with interruptions, accents, jargon and decisions. |
| Workplace chat and email | Workplace chat and email show how teams coordinate, decide and hand off work over months and years. |
| Software engineering data | Software engineering data shows how professional teams write, review and ship code: private repositories with full history, pull requests with review threads, issue trackers and incident records. |
| Sales and CRM data | Sales and CRM data records how deals actually progress: accounts, opportunities, activities, emails, proposals and the final won or lost outcome. |
| Company documents and knowledge | Company documents and knowledge bases hold how an organization actually works: SOPs, runbooks, wikis, and documents with their revision history. |
| Finance and accounting data | Finance and accounting data shows the work behind the numbers: invoice coding, approvals, reconciliations, close checklists and financial models. |
| Legal and compliance documents | Legal and compliance data covers how companies negotiate, approve and document obligations: contract drafts and redlines, signed agreements with clause data, and completed security questionnaires. |
| Operations and workflow data | Operations and workflow data records how work actually gets done: task histories with decisions and outcomes, field-service jobs, quality inspections, and recordings of people doing real work. |
All listings
Browse by use case
- Enterprise data for training AI agents
AI agents that do business work need examples of that work: what a person saw, what they did in which system, and how it turned out.
- Data for reinforcement learning environments and reward models
Reinforcement learning needs a way to tell good work from bad.
- Private data for model evaluation
Evaluations only measure what they claim if the model hasn’t seen the test data.
How requests work
Describe what you need
Domain, intended use, volume, format, history, geography, language and timeline, in the request form.
SourceX qualifies supplier companies
It identifies businesses that hold relevant data and can license it.
Scope and preparation are agreed
Permitted use, exclusivity, de-identification and delivery format are set with the supplier.
Data is delivered
After an executed agreement and the supplier’s authorization, with the agreed scope, permitted uses and preparation record.
The form also asks for a budget range, from under $50K to $1M+, and your licensing and de-identification requirements.
Availability
- On request
- SourceX sources this kind of data to order. It matches your request with companies that hold it, then agrees scope and preparation.
- Available
- Used once a specific package is under agreement and can be scoped for licensing. All current listings are on request.
Questions
Can I buy these datasets off the shelf?
All listings are currently sourced to order: SourceX sources the data from companies that hold it, then manages rights, scope, preparation and delivery. A listing is marked Available once a specific package is under agreement.
How is the data priced?
Per package. The request form asks for a budget range, from under $50K to $1M+. See how AI data deals are priced on sourcex.si.
Can I get exclusive rights?
Exclusivity can be agreed for a dataset snapshot.
How is personal information handled?
De-identification and redaction requirements are agreed with the supplier before any work begins. Each dataset comes with its agreed scope, permitted uses and a preparation record.
Does the data come with documentation of its rights?
Every dataset has a documented owner and confirmed licensing rights, and suppliers confirm their right to license the data before delivery.
Need data that isn’t listed?
Describe it in a request. SourceX looks for companies that hold it and can license it.