Senior AI Engineer Needed to Build Document RAG & Property Due Diligence Platform

I am looking for a senior AI engineer / full-stack developer to build a document intelligence platform for real estate, planning, legal, and property due diligence workflows. The goal is to let users upload property documents such as planning certificates, zoning reports, contracts, title documents, survey PDFs, council documents, and related reference files. The system should extract key information, validate it against external data sources, answer user questions with citations, and generate structured due diligence reports. This is not a basic PDF chatbot. I need a reliable document analysis system that can handle messy PDFs, scanned documents, tables, long reports, maps, and conflicting information across multiple files. Core requirements: * Upload and manage multiple PDF/document types * OCR for scanned or image-based PDFs * Text extraction from digital PDFs * Table and field extraction * Document classification by type * RAG pipeline with vector search * Cited answers with source document, page number, and extracted snippet * Property information extraction such as lot number, address, zoning, overlays, constraints, council references, dates, and conditions * Validation against external datasets or APIs where available * Conflict detection between documents * Risk scoring for each property or file * Automated due diligence report generation * Admin dashboard for document review, processing status, and extracted results * User dashboard for projects, properties, documents, chat, and reports * Background processing for large files * Audit trail so every AI answer can be traced back to the original source Preferred technical stack: Python, FastAPI, React or Next.js, PostgreSQL, pgvector or Qdrant, AWS Textract or Google Document AI, PyMuPDF, OpenAI or Claude, LangChain/LangGraph, AWS S3, Redis/Celery or similar queue system, Docker, and cloud deployment on AWS or GCP. The freelancer must clearly understand the difference between a simple PDF chatbot and a production document intelligence system. I need someone who can design extraction logic, OCR flow, chunking strategy, embeddings, citation handling, human review, report generation, and scalable backend architecture. Expected deliverables: * Technical architecture * Document upload and processing pipeline * OCR and extraction module * RAG-based Q&A with citations * Property/document data extraction * Risk scoring logic * Report generation * Frontend dashboard * Admin review interface * Deployment-ready backend * Documentation for future development Budget: $8,000–$25,000 Duration: 2–5 months Experience level: Expert Project type: Long-term product build with possible ongoing support Please apply only if you have experience with document AI, OCR, RAG, vector databases, structured extraction, PDF processing, and full-stack SaaS development. In your proposal, include examples of similar systems you have built and explain how you would approach OCR, chunking, citations, and report generation.