Python Developer for Image-Matching System

Python Developer: Automated Art Image-Matching System I need an experienced Python developer to build an automated image-matching system for provenance research. What it does: The system monitors ~480 international auction house websites for new listings. It compares each new listing against a private archive of ~80,000 artwork images, using OpenCLIP embeddings and a local Qdrant vector database. About this project: A detailed technical spec written by the user covers architecture, data flow, tech stack, and a 5-phase plan. It's available on request. Five components: Archive indexing — Google Drive + OpenCLIP + Tesseract OCR + Qdrant Web scraper — ~480 sites (static HTML, JS-rendered via Playwright, and anti-scraping-protected) Matching engine — cosine similarity + OCR text matching Email digest — HTML email via SendGrid, 3x/week Scheduling — Windows Task Scheduler, fully automated Runs on a local Windows PC with an Nvidia GPU. No cloud hosting. Looking for: Strong Python Real scraping experience (Requests, BeautifulSoup, Playwright) Hands-on computer vision / ML (PyTorch or similar) Familiarity with vector databases Comfortable with Windows deployment Nice to have: CLIP/OpenCLIP experience, Tesseract OCR, archival data projects. Logistics: Remote. Fixed-price. ~5 weeks. A small paid test task (one auction-house scraper) is available before committing to the full project.