PDF2Data – From Documents to Clean Datasets

Turning unstructured PDFs into analysis-ready data pipelines.

Roadmap

1. Phase 1 – Structured Extraction (MVP)
   • Upload PDFs → extract predefined fields / tables.
   • Support invoices, contracts, inspection reports.
   • Export CSV / JSON; basic preview UI.
2. Phase 2 – Data Cleansing Pipeline
   • Rule-based + LLM-assisted deduplication, type casting, currency/date normalisation.
   • Interactive cleaning UI with bulk actions & undo.
3. Phase 3 – Intelligent Analytics
   • Merge multi-file datasets, build KPIs and anomaly detection.
   • Natural-language querying (Chat over data).