AI Bid Database Enrichment Agent

How Cassidy automates bid intake and leveling using AI

Step 1: Trigger on new bid submission or import

The Workflow activates when bids arrive—via connectors to BuildingConnected, Procore, shared mailboxes, or drag-and-drop uploads—automatically deduplicating revisions and flagging addenda acknowledgment status.

Step 2: Parse and extract bid data

Cassidy applies OCR to scanned documents and uses layout-aware parsing to extract line items, quantities, unit prices, alternates, allowances, exclusions, and commercial terms from each submission.

Step 3: Normalize vendors and map cost codes

The agent resolves vendor name variants to your master list, attaches compliance data (EMR, insurance, bonding capacity), and maps every line item to your WBS or CSI MasterFormat structure for consistent categorization.

Step 4: Level bids and detect scope gaps

Cassidy aligns each bidder to your scope checklist, flags missing items or unacknowledged addenda, adjusts for inclusions and exclusions, and isolates alternates so base prices are directly comparable.

Step 5: Calculate risk scores and flag outliers

The Workflow computes weighted risk scores—factoring scope completeness, price variance against historical benchmarks, commercial compliance, and vendor performance—and highlights statistical outliers for estimator review.

Step 6: Generate deliverables and update historical database

Cassidy produces leveled bid tabs, risk scorecards, and benchmark reports exportable to Excel or PDF, then writes enriched records to your historical database with full data lineage for future estimating and continuous model improvement.