Spaces:
Running
Running
| title: Programme World Model | |
| emoji: 🧭 | |
| colorFrom: green | |
| colorTo: blue | |
| sdk: static | |
| app_file: index.html | |
| pinned: true | |
| short_description: Evidence-grounded AI for complex programmes | |
| # Programme World Model | |
| A research and engineering project by **Bruno Chikeka**, exploring | |
| evidence-grounded AI for complex programmes. | |
| This Space presents the project’s approach, current implementation and research | |
| limits. It is a static overview, not a trained forecasting model or an interactive | |
| simulation service. | |
| The current implementation includes a public-report evidence pipeline, a | |
| versioned SQLite index, resumable acquisition, source-review and reconciliation | |
| records, backup/restore, and a separate synthetic simulation kernel. | |
| The initial pilot covers 15 implementation reports for World Bank project | |
| P126507. It retains original document hashes and cited observations. Source | |
| coverage, historical availability and independent extraction accuracy remain | |
| under investigation. No model is empirically calibrated. | |
| Only this overview is published here. The implementation repository and source | |
| corpus are not included in this Space. | |
| Profile: https://huggingface.co/Chikeka | |