Clinical EMR Platform Codebase
Production-grade clinical platform codebase available for commercial AI training licensing.
Overview
This dataset contains the full source code of a production electronic medical records (EMR) and clinical management platform, built and deployed over 27 months with real doctors across 40+ medical specialties.
The codebase is available as a non-exclusive commercial license for AI model training, fine-tuning, and benchmarking purposes.
What Makes This Valuable for AI Training
Real engineering history — not synthetic code.
- 6,267 commits over 27 months of active development
- 2,381 pull requests averaging 203 lines of code each — real code review decisions, revisions, and approvals made under production conditions
- 18 unique contributors
- 2,565 source files across 31 repositories
Domain specificity. Specialty-specific clinical decision logic across 40+ medical specialties including psychiatry, obstetrics and gynecology, and pediatrics. Rule-based medication suggestion engines, emergency alert triggers, and patient intake flows built to real clinical standards.
Production validation. This code was deployed to 150+ paying doctors who used it for real clinical workflows. It has been tested against real-world cases
Technical Specifications
| Metric | Value |
|---|---|
| Total Lines of Code | 338,607 |
| Repositories | 31 |
| Commits | 6,267 |
| Pull Requests | 2,381 |
| Avg PR Lines | 203.6 |
| Unique Contributors | 18 |
| Development Period | 27 months |
| Source Files | 2,565 |
| Dependencies | 393 packages |
Language Composition
| Language | Lines | Share |
|---|---|---|
| JavaScript | 213,924 | 63.2% |
| Java | 87,979 | 26.0% |
| TypeScript | 34,004 | 10.0% |
| Python | 2,565 | 0.8% |
Stack
Spring Boot · React · Next.js · Java · TypeScript · Python · SQL
Use Cases
- Fine-tuning coding LLMs on domain-specific enterprise Java and React
- Training code review and code synthesis models using real PR history
- Benchmarking coding agents on clinical/healthcare domain code
- Building healthcare-specific coding assistants
Licensing
This dataset is available under a commercial non-exclusive license.
- Licensor retains full IP ownership
- Buyer receives perpetual non-exclusive rights for AI training use
- Multiple licenses available — non-exclusive means you are not the sole licensee
This dataset is gated. To request access, contact:
Please include your organisation name, intended use case, and estimated training scope in your request.
- Downloads last month
- 6