A B2B SaaS platform connecting brands with communities — built end-to-end as a monorepo with a real-time Convex backend, a hybrid AI search engine across 85,000 communities, Stripe partnership payments, and a Shopify affiliate app.

Brands running community-led growth face a research problem: there are tens of thousands of relevant online and offline communities — running clubs, wellness groups, professional networks, subcultures — but no structured way to find, evaluate, and partner with them. The process was manual, slow, and impossible to scale.

TRES was built to solve this end-to-end: discover communities by describing what you need in plain language, evaluate fit with AI-generated scores and reasoning, initiate partnerships with structured workflows and escrowed payments, measure the resulting social performance, and attribute revenue through an affiliate commerce layer. Coralsoft was responsible for the entire product engineering surface — architecture, Convex backend, all frontends, ingestion pipelines, LLM enrichment, Typesense search, Chrome extension, Shopify app, and admin tooling — evolved continuously as the corpus grew from 3,000 to 85,000 communities.