🚀 PitchPilot: AI-Powered Investor Deck Analyzer & Coach

Transform your pitch deck from good to fundable with AI-powered multi-agent analysis

🎯 What is PitchPilot?

PitchPilot is an enterprise-grade AI platform that helps founders perfect their investor pitch decks. Using a sophisticated multi-agent AI system, it provides comprehensive analysis, actionable feedback, and investor-perspective insights to maximize your chances of securing funding.

🤖 Multi-Agent Orchestration

Our LangGraph-powered system employs 3 specialized AI agents working in concert:

1. 🧠 Supervisor Agent - OpenAI-powered workflow orchestrator that intelligently routes tasks between agents
2. 📊 Analysis Agent - Deep-dives into your pitch structure, narrative flow, and completeness using VC frameworks
3. 💯 Scoring Agent - Evaluates across 5 key dimensions: clarity, differentiation, traction, scalability, and overall rating

✨ Core Features

For Founders

• Intelligent Analysis - Get VC-grade feedback in seconds, not weeks
• 📈 Comprehensive Scoring - Know exactly where your pitch stands (0-10 scale)
• 💬 Smart Workflow Routing - AI supervisor determines what analysis you need based on your query
• 📄 Multi-Format Support - Upload PDF, PPTX, DOCX, or TXT files
• 📊 Professional Dashboard - Modern UI to visualize your pitch performance

Technical Excellence

• ⚡ Real-time Processing - Async architecture for lightning-fast analysis
• 🔐 Enterprise Security - Secure file handling with Supabase storage
• 🎭 Smart Routing - LangGraph orchestrates agent workflows intelligently
• 📊 Structured Output - Type-safe responses with Pydantic validation

🏗️ Architecture Overview

graph LR
    A[Frontend - Next.js] --> B[FastAPI Backend]
    B --> C[LangGraph Orchestrator]
    C --> D[Supervisor Agent]
    D --> E[Analysis Agent]
    D --> F[Scoring Agent]
    E --> G[OpenAI]
    F --> G

Loading

🚀 Quick Start

Prerequisites

• Python 3.10+ & Node.js 16+
• PostgreSQL database
• OpenAI API key
• Supabase account (for storage)

1️⃣ Clone & Setup

# Clone the repository
git clone https://github.com/yourusername/PitchPilot-AI-Powered-Investor-Deck-Analyzer-Coach.git
cd PitchPilot-AI-Powered-Investor-Deck-Analyzer-Coach

# Setup Python environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r backend/requirements.txt

# Setup Node.js dependencies (using pnpm)
cd frontend
npm install -g pnpm
pnpm install

2️⃣ Environment Configuration

Create .env in the backend folder:

# Database
DATABASE_URL="postgresql://user:password@localhost:5432/pitchpilot"
DIRECT_URL="postgresql://user:password@localhost:5432/pitchpilot"

# OpenAI
OPENAI_API_KEY="sk-..."
OPENAI_MODEL="gpt-4.1-mini"
OPENAI_MODEL_SUPERVISOR="gpt-4.1"

# Supabase
SUPABASE_URL="https://xxx.supabase.co"
SUPABASE_KEY="xxx"

Create .env in the frontend folder

NEXT_PUBLIC_API_URL=http://localhost:8000
NEXT_PUBLIC_APP_NAME=PitchPilot

3️⃣ Database Setup

# Generate Prisma client
npx prisma generate

# Push schema to database
npx prisma db push

4️⃣ Launch the Platform

# Terminal 1: Start the backend
cd backend
python main.py

# Terminal 2: Start the frontend
cd frontend
pnpm run dev

🎉 Visit http://localhost:3000 to start analyzing!

📁 Project Structure

PitchPilot/
├── 🎨 frontend/                 # Next.js 15.3.3 React application
│   ├── src/
│   │   ├── app/                # App router pages
│   │   ├── components/         # React components
│   │   ├── lib/               # API client & utilities
│   │   └── types/             # TypeScript definitions
│   └── public/                # Static assets
│
├── 🧠 backend/                 # FastAPI + LangGraph backend
│   ├── app/
│   │   ├── ai/               # Multi-agent system
│   │   │   ├── agents.py     # Agent implementations
│   │   │   ├── pitch_graph.py # LangGraph workflows
│   │   │   └── config.py     # AI configuration
│   │   ├── api/              # API endpoints
│   │   │   ├── api.py        # FastAPI app configuration
│   │   │   └── routers/      # Route handlers
│   │   │       └── pitch_api.py
│   │   ├── config/           # Application configuration
│   │   │   ├── prisma_client.py
│   │   │   └── logging_config.py
│   │   ├── schemas/          # Pydantic models
│   │   │   └── pitch_schema.py
│   │   └── services/         # Business logic
│   │       ├── db_actions.py
│   │       ├── file_service.py
│   │       └── supabase_connection.py
│   └── main.py               # Application entry
│
└── 📊 prisma/                  # Database schema
    └── schema.prisma

🤖 The AI Agents Explained

🧠 Supervisor Agent

• OpenAI-powered workflow orchestrator
• Analyzes user queries to determine optimal agent routing
• Implements smart fallback logic for robust operation
• Routes between analysis and scoring based on user needs

🔍 Analysis Agent

Examines your pitch deck using VC frameworks for:

• Problem-solution fit clarity using Y Combinator standards
• Market opportunity presentation with a16z evaluation criteria
• Team credibility signals based on top-tier VC best practices
• Business model viability assessment

📊 Scoring Agent

Provides quantitative evaluation across 5 key dimensions:

• Clarity (0-10): How well you communicate your idea
• Differentiation (0-10): Your competitive advantage strength
• Traction (0-10): Proof of concept/growth evidence
• Scalability (0-10): Growth and expansion potential
• Overall (0-10): Comprehensive pitch strength

🎯 Q&A Simulation Agent

Generates mock investor questions from your deck:

• Coming Soon - Intelligent question generation based on your pitch content
• Simulates real investor meeting scenarios
• Helps you prepare for tough questions before the actual pitch
• Identifies potential weak spots that investors might probe

🛠️ Tech Stack

Frontend

• Framework: Next.js 15.3.3 with App Router
• UI: React 19 + Tailwind CSS 4
• Language: TypeScript 5
• Icons: Lucide React 0.513.0
• Package Manager: pnpm

Backend

• API: FastAPI 0.104.0+ with async/await
• AI Orchestration: LangGraph 0.0.20+ + LangChain 0.0.335+
• LLM: OpenAI 1.3.0+
• Database: PostgreSQL + Prisma ORM 0.10.0+
• Storage: Supabase 2.0.0+

Document Processing

• PDF: PyPDF2 3.0.1+ & pdfplumber 0.10.0+
• Word: python-docx 1.1.0+
• PowerPoint: python-pptx 0.6.23+
• Structured Output: instructor 1.0.0+

📚 API Reference

Main Endpoints

POST /evaluate-pitch

Upload and analyze a pitch deck

Request:

- file: PDF/PPTX/DOCX/TXT (max 10MB)
- title: string (pitch title)
- description: string (optional pitch description)
- user_query: string (AI analysis prompt - what you want analyzed)

Response:

{
  "feedback": {
    "overall_feedback": "Your pitch demonstrates...",
    "strengths": "Strong market opportunity...",
    "weaknesses": "Limited traction evidence...",
    "opportunities": "Consider expanding...",
    "threats": "Potential competitive risks...",
    "suggestions": "Prioritized improvement steps..."
  },
  "score": {
    "clarity": 8.5,
    "differentiation": 7.2,
    "traction": 6.8,
    "scalability": 8.0,
    "overall": 7.6
  }
}

🚦 Development Workflow

1. Local Development

   # Backend hot-reload
   cd backend && python main.py
   
   # Frontend with hot-reload
   cd frontend && pnpm run dev

2. Database Migrations

   npx prisma migrate dev --name your_migration_name

3. Type Generation

   npx prisma generate

🙏 Acknowledgments

Built with ❤️ for founders by founders, powered by:

• LangGraph for agent orchestration
• OpenAI for language models
• Next.js for the frontend framework
• FastAPI for the backend API

---

Ready to perfect your pitch?
🚀 Start Analyzing Now