Forge BI
AI / Fintech
Overview
Forge BI is an AI-powered business intelligence platform designed specifically for startups. It connects to your financial data sources — bank accounts, accounting software — and uses natural language AI to deliver actionable insights.
Built with Next.js 15, TypeScript, and Vercel AI SDK — integrates Plaid for bank data and applies 8 years of financial analysis experience to product design, creating dashboards that translate raw data into decisions.
Challenge
Startups need financial visibility but can't afford dedicated analysts. Existing BI tools are complex, expensive, and require SQL knowledge. The goal was to create a platform that makes financial intelligence accessible to non-technical founders through conversational AI.
Tech Stack
Architecture
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. 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.
Key Features
AI Chat
Natural language interface for querying financial data. Ask questions in plain English and get charts, tables, and insights.
Bank Sync
Plaid integration for automatic bank account syncing. Real-time transaction data flows into the platform without manual entry.
Goals
Auto-tracked financial goals — revenue targets, profit margins, break-even analysis. Progress updates automatically from live data.
What I Learned
- Domain knowledge changes the schema. A general-purpose finance schema would have made the AI's job harder — naming things the way analysts name them (variance, runway, burn) lets the model reason in the user's language.
- AI as an analyst replacement, not a chatbot. The interface prefers charts and tables over conversation; chat is the input, not the output.
- Plaid's realtime story is rougher than the marketing implies. Webhook reconciliation + a backfill cron is more reliable than trusting any single signal.