Holding company / Data lab
We acquire the data that AI is trained on.
Monterra Holdings sources, acquires, and refines proprietary datasets from real-world businesses, then licenses them to teams building frontier and domain-specific AI models.
Approach
A holdco built around one asset class: data.
Acquire
We buy datasets outright, acquire companies for the data they generate, and structure revenue-share partnerships with operators sitting on valuable records.
Refine
Our data lab cleans, de-identifies, labels, and documents every corpus so it is ready for pre-training, fine-tuning, and evaluation.
License
We license curated datasets to AI labs and enterprises under clear terms — exclusive or non-exclusive, one-time or continuously refreshed.
Portfolio
Monterra AI
Our operating company. Monterra AI deploys MIRA, an AI operating system for multi-entity enterprises, and helps portfolio companies productize and license their operational data — feeding the Monterra Holdings data pipeline.
Standards
Data you can actually train on — and defend.
- Documented provenance
- Every record traces back to a source with clear rights to sell or license it.
- Privacy by default
- Personal information is removed or de-identified before any dataset leaves the lab.
- Clean licensing
- Straightforward commercial terms that hold up to legal and procurement review.
- Model-ready format
- Structured, labeled, and delivered with datasheets describing scope and limitations.
Contact
Buying data, or sitting on it?
AI teams looking for proprietary training data, and businesses or founders with datasets to sell — we'd like to hear from you.
team@monterraholdings.com