AI Computer Vision Engineer Needed for Blueprint Fire Door Detection (SaaS Potential)

We are a fire door inspection company looking to develop an AI-powered application that can analyze architectural blueprint PDFs and automatically identify fire-rated door openings. Project Overview The goal is to create an MVP that allows a user to upload a set of architectural plans and receive an accurate inventory of all fire-rated doors within the building. The system should be capable of: Reading architectural blueprint PDFs. Detecting and identifying door symbols. Reading door tags and door schedules. Cross-referencing door schedules with the floor plans. Identifying which doors are fire-rated based on schedules, legends, wall ratings, or other available plan information. Highlighting identified fire doors directly on the drawings. Exporting a complete door list to Excel or CSV. Preferred Experience We’re looking for someone with experience in one or more of the following: Computer Vision OCR (Optical Character Recognition) Architectural or construction blueprint analysis PDF parsing Python OpenCV YOLO or similar object detection models AI document analysis Experience working with architectural drawings, engineering plans, construction documents, or similar technical drawings is strongly preferred. Deliverables The initial MVP should include: Blueprint PDF upload Automated fire door detection Highlighted output drawing Door inventory export (Excel/CSV) Clean, documented source code Future Opportunity This is intended to become much more than an internal tool. The long-term vision is to build a commercial SaaS platform for the fire door inspection industry. The right developer could have the opportunity to continue working on future phases, including additional AI features, cloud deployment, customer accounts, and automation tools. When Applying Please include: Examples of projects involving computer vision, blueprint analysis, or document AI. A brief explanation of how you would approach this project. Your recommended technology stack. An estimated timeline for an MVP. Any questions or suggestions you have after reading this description. We’re looking for someone who enjoys solving difficult technical problems and is interested in building a product with long-term potential—not just completing a one-time freelance project.