Direct data licensing
Patronus AI data partnerships
Patronus AI buys licenses for private codebases and company records used in AI training and evaluation work. Its process separates an initial inquiry, sample review and full-delivery acceptance. A separate partner program compensates introductions; this is not an individual task-work marketplace.
- Best suited to
- Owners of private codebases and connected business records
- Published starting point
- Rights holder or authorized company representative; review applies
- Revenue model
- Direct license fee after acceptance
- Main decision
- Whether the proposed dataset and permitted uses match your rights
Jump to a section
Who it’s for
Who is Patronus AI’s data program for?
Patronus identifies code, company records and released-game material as licensing routes. Its partner page emphasizes development history and records that connect activity across a business. It states that Patronus is the buyer.
Our fit test: can you describe a coherent workflow rather than a disconnected collection of files? A small, clearly scoped inventory helps an owner explain the opportunity without prematurely exporting sensitive records.
Get prepared
What should you prepare?
Start with a short inventory and an ownership check. The following preparation steps are editorial guidance, not additional provider eligibility rules.
- Asset summary: Describe the software or workflow, dates covered and systems involved.
- Decision authority: Identify the owner and the person authorized to discuss licensing.
- Sample boundary: Choose a representative slice only after agreeing on confidentiality, privacy and evaluation terms.
- Acceptance questions: Ask how delivery will be checked, what constitutes rejection and how exceptions are resolved.
Use our data preparation guide to organize the inventory before sharing sensitive material.
Revenue & payments
How does payment work?
The owner page describes a sample under NDA, a written offer and up to 30 days to review full delivery. Payment follows acceptance directly from Patronus, rather than waiting for another buyer. No universal owner price is published.
The partner program advertises 15% of each introduced license fee after the owner is paid, subject to a written partner agreement. This is introduction compensation, not an owner revenue-share percentage.
Data Mining Masters links directly without tracking. Your referral URL is still needed; publication does not mean enrollment or an approved partnership.
Rights & control
Which licensing terms deserve attention?
Patronus describes non-exclusive licensing and says it uses material to build training and evaluation environments, rather than reselling the original records themselves. The agreement defines permitted use and sharing. Its page says rejected material is deleted with written confirmation.
Keep a copy of the approved scope and review criteria. For records spanning several departments, map who can authorize each system and identify contractual restrictions before offering a sample.
- Who can receive the licensed material, and for which purposes?
- What happens to existing licenses after withdrawal or termination?
- Who resolves third-party rights, privacy and confidentiality restrictions?
- Which acceptance criteria, deductions and payment deadlines will be written into the agreement?
Explore your options
When should you compare another platform?
Compare Pangea.ai for code-focused licensing, or Handshake AI for another business-data route. If you want paid tasks rather than licensing an asset, start with our AI-work guide.
Good questions, clear answers
Questions before you apply
Do I need to export everything first?
No. Start with a description. Agree on the sample scope and terms before transferring records.
Is an introduction enough to earn a fee?
The advertised partner payment follows a completed license and owner payment. An introduction alone is not a guaranteed commission.
How should I compare a written offer?
Compare the asset scope, allowed uses, review criteria, payment trigger and termination terms together. A headline amount alone is not enough.
Sources and editorial approach
Provider information checked on 18 November 2026. Practical checklists are Data Mining Masters’ editorial guidance. This profile is based on public information, not a firsthand transaction or a guarantee of acceptance or income. How we compare platforms.
Keep exploring