Document AI
SEC Filings Table Extraction – Real-Time & 2006–2026 Historical Reprocessing
Semantic financial-table JSON from live SEC filings and the 2006–2026 archive.
Project Overview
Built an HTML-table extraction pipeline for SEC filings (10-K, 10-Q, 8-K, 6-K, proxy statements) that replaces each source table with a hash, rebuilds the grid (colspan/rowspan, multi-row headers, merged currency and % columns), detects the nearest table title and converts it to a lossless semantic JSON with row and column tags for a client-facing product. Added suppression rules for junk tables and fixed data-loss bugs, then reprocessed the full 2006–2026 archive across multiple servers using a local database synced back to production.
Tools & Technologies: PHP · Node.js · HTML/DOM parsing · Python · MySQL · JSON · Excel
Semantic table output
Merged-cell handling
Header preservation
Historical reprocessing
Project capabilities are described from recorded experience. Artwork uses illustrative sample data.
Need a similar system?
Let’s discuss how I can help build a solution around your documents, data, and workflows.
More to explore
All projects →
Global Filings PDF Parser
Real-time text, headings, and financial-table extraction from global filings.
AWS Announcements Pipeline
Parallel serverless PDF parsing for a stock-exchange announcements feed.
Extracted-Table Review Portal
Source-linked PDF comparisons, table review views, and structured-output inspection.