Infra for Government Fraud Hunters

Build an AI-native fraud case-building platform that accelerates False Claims Act (FCA) and qui tam investigations for whistleblower law firms, state AGs, and inspectors general. Objective: Transform a raw insider tip into a structured, evidence-backed, complaint-ready case — reducing investigation time from months or years to weeks. This is not a fraud dashboard. It is an intelligent system that ingests messy documents, traces opaque corporate relationships, extracts patterns of misconduct, and assembles legally defensible case files. Core vision: Today, fraud investigations under the False Claims Act are slow and manual: - An insider submits a tip - Lawyers gather documents - Analysts sift through PDFs and claims data - Corporate structures are manually mapped - Evidence is assembled by hand This platform uses AI to automate the evidence organization and narrative-building process while preserving legal traceability. Core system architecture: 1. Tip-to-Case Intake Engine - Secure whistleblower submission portal - Structured intake for allegations, timelines, and known actors - Initial AI-assisted hypothesis extraction - Automatic case workspace creation Goal: convert a narrative tip into structured investigative hypotheses. 2. Document Ingestion and Parsing Pipeline - Upload and parse large volumes of PDFs, emails, contracts, claims data, and spreadsheets - OCR for scanned documents - Table extraction for invoices and billing records - Entity extraction (names, NPIs, EINs, addresses, bank accounts, contract IDs) - Page-level citation tracking for legal defensibility All extracted facts must link back to source location. 3. Entity Resolution and Corporate Structure Mapping - Resolve related entities across datasets - Detect shared addresses, ownership, management, and payment flows - Map shell companies and layered corporate hierarchies - Build an explainable relationship graph Investigators should see how entities connect and where suspicious overlap occurs. 4. Pattern Detection and Evidence Clustering - Identify anomalous billing patterns - Flag duplicate claims, suspicious timing, and inflated charges - Compare against peer benchmarks - Surface statistically abnormal behavior with explainable factors Each finding must include: - Supporting documents - Confidence score - Clear reasoning 5. AI Case Builder - Organize findings into structured allegations - Generate draft complaint sections with citations - Build timeline of misconduct - Assemble exhibit index automatically - Package evidence into complaint-ready files Every claim must be traceable to underlying evidence. 6. Investigator Workbench - Case timeline with immutable action log - Evidence graph visualization - Task assignments and SLA tracking - Notes and annotations with citation support - Referral workflow for DOJ or AG escalation 7. Legal-Grade Audit and Traceability - Immutable investigation timeline - Versioned narrative drafts - Source citation tracking (document, page, snippet) - Exportable audit logs - Oversight-ready reporting Roles: - system-admin - fraud-manager - investigator - analyst - oversight-auditor Data model: - tips - programs - transactions - entities - entity-links - parsed-documents - evidence-snippets - fraud-patterns - investigation-cases - allegations - exhibits - case-actions - referrals - recoveries - audit-events Performance metrics: - Time from tip to complaint draft - Average documents processed per case - Evidence coverage ratio - Case cycle time - Dollars recovered per investigation - False-positive trend - Referral acceptance rate System requirements: - Secure data isolation and encryption - Strong access controls - Full traceability for legal proceedings - High-volume document processing - Scalable entity graph engine - Explainable AI outputs Create a modern startup design inspired by Y Combinator (YC) companies. Choose one bright primary color and build a clean, minimal color scheme around it. The design should feel bold, simple, and product-focused with strong typography, generous whitespace, and clear hierarchy.