
Forge BI
AI-powered business intelligence for startups. Financial insights via natural language.
Plaid pulls transaction data into a Prisma-modeled financial schema (accounts, transactions, derived metrics). The AI Chat layer is a Vercel AI SDK orchestration that translates natural-language questions into typed queries over that schema, then renders charts/tables inline. Goals are background-evaluated against the same data, no manual reconciliation step. Plaid sync runs from two directions: a signature-verified webhook fires an incremental transactionsSync the moment a bank posts an update, and a twice-daily cron re-runs the same sync across every active connection as a safety net, both paging through a stored cursor and upserting on Plaid's transaction id so retries never double-count. Eight years of healthcare finance taught me which questions founders actually ask, so the schema and prompt scaffolding are tuned for variance analysis and runway over generic BI.
- 1 schema, 3 surfacesPlaid bank sync, AI chat, and auto-tracked goals all share one Prisma data model. No reporting DB, no ETL.
- 8 yrs domain DNATables named in analyst language (variance, runway, burn) so the AI reasons in the user's vocabulary, not generic BI.





