653 pages in. A printable risk brief, a priced material takeoff, and verbatim spec citations out — by Monday. If it earns its keep, $249 next time. If not, no follow-up, no hard feelings.
Your bid PDFs are never used to train models. End-to-end encryption. Delete on request.
“all work shall be done so as not to interrupt or interfere with the daily/occupancy of the building. this shall include the provision of a temporary fuel supply if necessary.”
“the addendum date controls; if you submit on the bid-form date your bid will be rejected as late.”
“all work shall be done so as not to interrupt or interfere with the daily/occupancy of the building. this shall include the provision of a temporary fuel supply if necessary.”
“the addendum date controls; if you submit on the bid-form date your bid will be rejected as late.”
BidLedger is a citation-verified construction bid analysis pipeline. You drop in a bid set — 600+ pages of drawings, project manual, addenda, bid form. It reads every page, finds risks and scope items, checks every finding word-for-word against the source PDF, and writes a printable risk brief plus a five-sheet priced material takeoff. Findings the AI can't verify never reach the report — they're logged separately so you can audit them. Built for GC chief estimators and MEP subs.
Five steps in the pipeline. Click any to see exactly what happens inside.
We hash every PDF, classify drawings vs. spec vs. addenda vs. bid form, extract text deterministically, and rasterize each page for the multimodal stage. Re-runs are cheap and idempotent.
612 pages · 8 documents · 4 addenda · 87 verified findings
“All hangers and supports for piping 2-1/2 inches and larger shall be furnished with vibration isolators per Section 23 05 48. Failure to comply shall result in field rework after structural inspection.”
Each event hash includes the previous event's hash. Change one event and the whole chain breaks visibly.
AI findings that failed citation verification are kept in findings-rejected.jsonl. Forensic trail intact.
Never used to train a model. Run inside our managed environment or self-host the pipeline on your infrastructure.
Email it over, or paste a shared link. Read by an estimator within 1 business day. If it earns its keep, $249 next time.