M13 Meeting Prep — AI Briefing Generator

Generates structured investor briefing documents ahead of meetings by pulling from live web search, CRM notes, and M13 thesis context.

Stack

• Backend: FastAPI + Python (Tavily, OpenAI, FAISS, Airtable)
• Frontend: React + Vite (react-markdown)

Setup

1. Clone and configure environment

cd backend
cp .env.example .env
# Fill in your API keys in .env

Required keys:

Key	Where to get it
OPENAI_API_KEY	platform.openai.com
TAVILY_API_KEY	app.tavily.com
AIRTABLE_API_KEY	airtable.com/create/tokens
AIRTABLE_BASE_ID	Your base URL: airtable.com/appXXXXX/...

Note: Airtable is optional for the demo — if not configured, the app uses realistic mock CRM data automatically.

2. Start the backend

cd backend
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

3. Start the frontend

cd frontend
npm install
npm run dev
# Open http://localhost:5173

Airtable CRM Schema

Create a table called Deals with these fields:

Field	Type
Company	Single line text
Stage	Single select
Notes	Long text
LastContact	Date
AssignedTo	Single line text
Tags	Multiple select

How it works

1. User enters company name + meeting context
2. FastAPI receives the request and kicks off 3 parallel tasks:
   • Tavily: 3 concurrent searches (recent news, funding, founder background)
   • Airtable: looks up prior deal notes by company name
   • FAISS RAG: retrieves relevant M13 thesis/portfolio context via semantic search
3. All results feed into a GPT-4o prompt
4. Response streams back token-by-token via SSE
5. User can copy markdown or download a styled PDF

Project structure

meeting-prep/
├── backend/
│   ├── main.py           # FastAPI app + endpoints
│   ├── orchestrator.py   # Parallel data gathering + GPT-4o streaming
│   ├── tavily_client.py  # Live web search
│   ├── airtable_client.py # CRM notes
│   ├── rag.py            # FAISS index over M13 corpus
│   ├── prompts.py        # Prompt templates
│   ├── pdf_generator.py  # Markdown → PDF
│   └── requirements.txt
└── frontend/
    ├── src/
    │   ├── App.jsx       # Main UI
    │   └── main.jsx
    └── package.json

Extending

• Add more corpus: Edit M13_CORPUS in rag.py with additional M13 portfolio/thesis content. Scrape m13.co/portfolio periodically to keep it fresh.
• Add Crunchbase: Swap the Tavily funding search for the Crunchbase API to get cleaner structured funding data.
• Slack integration: Post the generated brief to the relevant deal channel automatically.