The AI cleaned SEC filing data for more than 600 companies after an audit found most split adjustments wrong
A man gave the AI strict tag research tasks and had it work through the whole ticker list again and again, searching and comparing XBRL tags in SEC filings. The pipeline now pulls 10-K and 10-Q filings for more than 600 companies, maps tags onto about 35 financial concepts, and produces cleaned data. Doing 3 tickers by hand had taken him about 3 weeks and still gave wrong data.
How it works
- He sets strict, no-regression tag research tasks
- The AI searches and compares XBRL tags across the full ticker list, company by company
- The pipeline pulls filings for more than 600 companies and maps tags onto about 35 concepts
- It works around filings that tag a full year as a quarter, cumulative cash flow, missing fourth quarters, changed tags, and stock splits
- It produces cleaned figures with prices added
Why it’s worth knowing
The audit behind the post found 82.7 percent of split adjustments in his old data were wrong, and the fix was an agent rechecking tags across hundreds of companies one by one.
Where it’s from
Summarized from a public post on news.ycombinator.com. See the original ↗