How buyer-match works

Buyer-match scores every active buyer in your pipeline against new listings as they come in. For each new listing, it returns the top 5 buyer fits with a confidence breakdown across price, location radius, bed/bath count, square footage, and learned search behavior. Average match completes in under 2 seconds per listing.

What does the realtor pilot cost?

The pilot is free for the first 14 days. No credit card required. Production pricing is published when the pilot exits — invite-only access during the early phase keeps support quality high.

Which MLS feeds are supported?

The pilot ships with public listing sources: Zillow, Redfin, and Craigslist FSBO. Live MLS / IDX integration is a paid add-on gated behind a feature flag and requires the realtor's MLS membership credentials. Title and tax records via DataTree or PropertyRadar are similarly opt-in.

How is buyer data kept private?

Each tenant runs in an isolated container with its own Supabase row-level-security scope. Buyer PII is redacted before any data leaves the container for embedding generation on the data layer. No tenant ever sees another tenant's data, and the agent's tool calls are scoped at JWT level, not application level.

What happens when the AI gets a match wrong?

Every recommendation is logged with a verdict (OFFER / FAIR / SKIP). Realtor feedback updates the per-tenant scoring weights. The system learns the realtor's actual preferences over time — not a generic average across all users.