# T2R2

> T2R2 is a platform for regulated software in pharma and healthcare. It builds GxP and medical-device (SaMD) applications fitted to the customer's process, and produces their regulatory evidence at the same time. Its agentic AI engine does the work; the customer's named experts review, decide and sign. T2R2 has been built since 2022. T2R2 is a registered trade mark of LLAAMA SAS (France).

Key facts, stated precisely:

- T2R2 stands for Transparent, Traceable, Replicable and Reproducible.
- The main idea: software and compliance used to be two projects run in parallel. T2R2 makes them one piece of work, so the work is done once, not twice.
- Fit for purpose: suites are bought whole and used in part, and the gaps end up in spreadsheets. T2R2 builds the application a process needs, works next to the systems the customer keeps (for example laboratory results and the electronic health record), and keeps it compliant as it changes. It does not ask the customer to replace those systems.
- T2R2 does three things for an application: the platform builds it (a linked chain from intended purpose to release, with human approval gates and an Evidence Pack with every release); the T2R2 library runs inside it (a tamper-evident record, replay of any past state, erasure that keeps the audit trail whole); the agentic AI harness keeps its AI agents under control (permissions, a human decision before anything changes a patient's care, replay of every model exchange, pinned model versions).
- Validation and certification belong to each application and to the organisation that owns it. T2R2 delivers the evidence; the customer's team reviews and signs it. T2R2 does not claim that any application is validated, certified or audited.
- Standards: evidence for EU MDR, IEC 62304, ISO 14971, IEC 62366-1, IEC 81001-5-1, the EU AI Act and ISO 13485; GxP applications built for the expectations of GAMP 5, 21 CFR Part 11 and EU Annex 11.
- Who it is for: quality and validation leaders at drug manufacturers and CDMOs; clinical-stage biotechs; hospitals that bring clinical applications into routine care; companies that build software as a medical device; teams in pharma and healthcare that deploy agentic AI and must explain what it does.
- Offer: an early-adopter pilot on one application, in one fixed-length cycle, at a fixed price, with exit criteria agreed at the start.
- Ownership: the customer owns the application, its documentation and its data, and can run it in its own data center. The T2R2 core is being prepared for open-source release under the Apache 2.0 licence.

## Pages

- [Home](https://www.t2r2.ai/): what T2R2 does, how it works, the Evidence Pack, recent examples, the pilot.
- [About T2R2](https://www.t2r2.ai/t2r2): the platform, the library, the agentic AI harness, the standards, the history since 2022, and frequently asked questions.
- [Full text for language models](https://www.t2r2.ai/llms-full.txt)

## Contact

- [Request a pilot](https://www.t2r2.ai/#contact) or write to weeksNotYears@t2r2.ai

## Optional

- [Legal notice](https://www.t2r2.ai/legal)
- [Privacy](https://www.t2r2.ai/privacy)
- [LLAAMA, the parent company](https://llaama.com)