Full-Stack AI Engineer — build an LLM-powered analytics MVP (~2-week build, likely ongoing)

Full-Stack AI Engineer — build an LLM-powered analytics MVP (~2-week build, likely ongoing) We're building a tool that measures how often, and how favorably, AI assistants (ChatGPT, Gemini, Claude, Perplexity) recommend a brand's products — then turns that into a score and a prioritized list of fixes. Think "search-rank tracking, but for the AI era." We need a pragmatic full-stack engineer to build the v1 fast. What you'll build: A query runner that calls multiple LLM APIs across a set of prompts, with repeat sampling An LLM-based extraction layer that parses each response into structured signals (is the brand mentioned? recommended? its rank? sentiment? which competitors? cited sources?) A scoring + recommendations layer A clean report UI (web + PDF) with background-job handling for the API runs Must have: Shipped LLM-powered products in production (not just demos) Strong with LLM APIs + prompt/structured-output design Next.js/React + Node or Python (FastAPI); Postgres/Supabase Async jobs/queues; comfort managing API cost and rate limits Nice to have: built an LLM-evaluation / "LLM-as-judge" system; data-pipeline experience; Stripe. Engagement: Contract, ~2–4 weeks for v1 (hourly or fixed milestone), strong potential for ongoing work. Start ASAP. Shortlisted candidates complete a short paid test task. Begin your reply with the word READINESS so I know you read this.