Agents for Good

Nonprofits spend enormous time on grant writing — researching funders, parsing guidelines, and drafting applications from scratch. We built a platform that automates this entire workflow. The system ingests uploaded grant guidelines (PDF/DOCX), uses AI to extract requirements and evaluation criteria, then generates tailored grant narratives using the organization's own data and voice. A semantic knowledge base powered by pgvector lets the AI draw on previously submitted grants and organizational documents for context.

A complete grant writing workflow: research potential funders, upload their guidelines, and the AI drafts a tailored application pulling from your organization's knowledge base. The question bank tracks common grant questions across funders so answers can be reused and refined over time. Built on Supabase with Row-Level Security for multi-tenant isolation, so multiple nonprofits can use the platform independently.