AI Backend & Architecture for Jersey Collector Community Platform

---

---

We’re building a next-generation community and marketplace app for authenticated sports jerseys—think Reddit/Discord social features + Instagram-style photos + seamless listing/transaction flows tailored to collectors. What You’ll Be Working On Community Verification & Categorization System Design a crowdsourced “legit-check” engine where users flag jersey authenticity (e.g. neckline stitching, label UPC, emblem details). Model & store contributor feedback at brand, model, team and player levels. Expose moderation insights (tips, decision rationale) for both users and admin moderators. “Quick-Add” from Photo Build an LLM-driven pipeline that ingests 2–3 smartphone photos and returns structured jersey metadata (brand, model year, player, etc.). Expose API endpoints for frontend to prefill inventory and marketplace listing forms. Personalized Feed Population Architect a basic recommendation engine that surfaces posts, listings and community content based on user history, watchlist/interests and engagement. Image Background Remover & Color Selector (optional, budget-permitting) Stand up a microservice to mask-out backgrounds and auto-suggest a solid backdrop that complements each jersey’s color palette. Pricing Data Insights Design data model for time series of jersey sale prices. Integrate an AI routine to detect and annotate value inflection points (e.g. “Spiked after Ovechkin broke the all-time goals record”). Ideal Skills & Experience AI/ML & LLM Integration: Demonstrable experience integrating large-language models and custom ML pipelines Computer Vision: Background removal, image metadata extraction, or similar Backend Architecture: Proven track record designing scalable APIs & data models (Node.js, Python, or similar) Datastores & Search: SQL/NoSQL modeling, full-text or vector search for feed and community queries Recommendation Systems: Basic collaborative/filter-based feed algorithms Cloud & DevOps: AWS/GCP/Azure service orchestration, Docker/Kubernetes a plus Collaboration: Strong communicator comfortable guiding a frontend engineer and documenting architecture Budget & Timeline Rough Allocation: Community Verification: ~$10K Photo “Quick-Add”: ~$3K Feed Population: ~$3K Background Tools: ~$1K Pricing Insights: ~$1K Duration: ~4–6 weeks To Apply Please share: -High-level thoughts on how you’d tackle our core community verification feature -Written bid and proposal based on the project We’d love to see your expertise and collaborate on building something truly unique for jersey collectors. Looking forward to your proposal!

Keyword: Cloud

Price: $17.0

Computer Vision Image Processing Machine Learning Sports & Recreation