Artist Opportunity Matching AI Collaboration

I’m looking for a research-minded AI team that can turn my existing dataset of artist profiles and historical matches into a robust system that automatically pairs creatives with fresh opportunities posted online.

The core of the work is experimental: I want to see how far large-language-model families such as Claude, Gemini and Grok can be pushed for nuanced profile-to-opportunity matching. That means going beyond simple keyword overlap and exploring richer text analysis, conversational reasoning and any clever prompt-engineering or fine-tuning tricks you can bring to the table.

You’ll start with several thousand annotated examples I already have. From there, I’d like you to design an approach, build a proof-of-concept pipeline, and iterate until we can clearly demonstrate higher match quality than the current rule-based baseline.

Key deliverables
• A documented research plan outlining model choice, data preprocessing and evaluation strategy
• Working prototype (scripts, notebooks or an API) that ingests new postings and ranks the best-fit artists
• Quantitative report comparing performance against baseline, plus recommendations for production hardening

If you’re comfortable experimenting with the latest LLMs, enjoy rapid iteration, and can explain your findings in plain English, I’d love to collaborate.

This is a weekly hire at $50 per week.