FundingAI — Funding Application Automation Tool

Automate startup funding applications by combining a company knowledge base with AI-powered content generation. Upload company docs once, paste a funding URL, and the system generates all submission materials.

Features

• Company Knowledge Base — Upload PDFs, DOCX, TXT, paste text, or import links. Content is chunked and embedded into a local vector store (ChromaDB) for RAG retrieval.
• Funding URL Scraper — Paste any funding page URL. AI extracts questions, deadlines, eligibility, and requirements into structured data.
• AI Generation Engine — Generates form answers, proposal narratives, executive summaries, and budget justifications with tone adapted per funder type (YC-style, government grants, VC, etc.).
• Pitch Deck Generator — Creates slide content (Problem → Solution → Market → Traction → Team → Ask) and exports to PowerPoint.
• Video Script Generator — Produces timed talking points tied to deck slides.
• Document Export — Word (.docx), PDF, and PowerPoint (.pptx) with styled formatting.
• Real-time Streaming — SSE streaming from backend to frontend for live generation output.

Tech Stack

Layer	Technology
Frontend	Next.js 14 (App Router), Tailwind CSS, TypeScript
Backend	Python FastAPI, SQLAlchemy, SQLite
AI	Claude API (Sonnet for short answers, Opus for narratives)
Vector Store	ChromaDB with local sentence-transformers embeddings
Scraping	httpx, BeautifulSoup, Playwright (fallback)
Export	python-docx, WeasyPrint, python-pptx

Architecture

┌──────────────┐     ┌──────────────────────────────────────────┐
│   Next.js    │────▶│  FastAPI Backend                          │
│   Frontend   │◀────│                                          │
└──────────────┘ SSE │  ┌─────────┐  ┌──────────┐  ┌────────┐  │
                     │  │ Scraper │  │ RAG/KB   │  │ Claude │  │
                     │  │ Service │  │ Service  │  │ API    │  │
                     │  └─────────┘  └──────────┘  └────────┘  │
                     │       │            │                     │
                     │  ┌─────────┐  ┌──────────┐              │
                     │  │httpx/BS4│  │ ChromaDB │              │
                     │  │         │  │ + SQLite │              │
                     │  └─────────┘  └──────────┘              │
                     └──────────────────────────────────────────┘

Quick Start

Backend

cd backend
cp .env.example .env   # Add your ANTHROPIC_API_KEY
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

Frontend

cd frontend
npm install
npm run dev

The Next.js dev server proxies /api/* requests to the FastAPI backend at localhost:8000.

Open http://localhost:3000 to use the app.

Usage

1. Upload company docs — Go to Company Profile, upload PDFs/DOCX/TXT or paste text. These get chunked and embedded for RAG.
2. Paste a funding URL — Go to Applications, enter a funding page URL. The scraper + AI extracts questions and metadata.
3. Generate materials — Open the application, use "Generate All" or generate individual sections (answers, narrative, deck, script).
4. Review & edit — Edit any generated content inline with the built-in editor.
5. Export — Download as Word, PDF, or PowerPoint from the Export tab.

API Endpoints

POST /api/profile/upload          Upload company files
POST /api/profile/text            Paste raw text
POST /api/profile/link            Import from URL
POST /api/scraper/extract         Scrape funding URL → structured requirements
POST /api/generate/answers/{id}   Generate form answers (SSE)
POST /api/generate/narrative/{id} Generate proposal narrative (SSE)
POST /api/generate/deck/{id}      Generate pitch deck (SSE)
POST /api/generate/all/{id}       Generate everything (SSE)
GET  /api/export/docx/{id}/{type} Download Word doc
GET  /api/export/pdf/{id}/{type}  Download PDF
GET  /api/export/pptx/{id}        Download pitch deck

Configuration

Set these in backend/.env:

Variable	Description	Default
ANTHROPIC_API_KEY	Your Claude API key	(required)
DATABASE_URL	SQLite connection string	sqlite:///./data/app.db
FRONTEND_URL	Frontend origin for CORS	http://localhost:3000
DEBUG	Enable debug logging	true