Skip to content

Repository files navigation

Deck to Insights - Automated Pitch Deck Analyzer

This project was developed as part of a technical round assessment.
It is an AI-powered due diligence pipeline that takes a startup's pitch deck (PDF/PPT), performs web research, validates key claims, and generates a structured investment report for early-stage investors.

Note: Developed my first agent! And first ever multi-agent orchestrated pipeline. Pulled an all-nighter for this because it was too interesting, spent a lot of time navigating through a couple challenges with agent prompting, 202 site blockage errors (which I did not realize was due to iteratively testing my web scrapers) and I could NOT find any startup pitch decks (yes even with Google search hacks).


Overview

Goal

To build a Python app/script that:

  • Ingests pitch decks in PDF/PPT format
  • Extracts key information and claims
  • Conducts external research beyond the deck
  • Validates startup claims using multi-agent reasoning
  • Outputs a well-structured Markdown report for an Investment Manager

Target User

Investment managers or analysts evaluating early-stage startups.


Architecture

The system uses three coordinated AI agents, each with distinct roles:

Agent Description
Orchestrator Agent Analyzes the deck, extracts company data, identifies gaps, and generates a research plan
Research Agent Conducts parallel DuckDuckGo searches, scrapes content (via Selenium/Requests), and summarizes findings
Validator Agent Fact-checks and verifies startup claims using structured evidence reasoning

Usage

pip install requirements.txt

python pipeline.py <.pptx or .pdf file>

Pipeline Flow

Input: Pitch deck (PDF or PPTX)
   ↓
Orchestrator → Extracts company info and creates research plan
   ↓
Research Agent → Executes searches, scrapes evidence
   ↓
Validator Agent → Validates claims using evidence
   ↓
Output: Markdown investment report  
    ├── *_deck_analysis.json
    ├── *_research_results.json
    ├── *_validation_report.json
    └── *_FINAL_REPORT.md           # Main deliverable

Future Enhancements

Slack Integration

  • Add Slack bot, easy to use for Investment Manager
  • Respond with markdown report in thread

Other Enhancements

  • Support for more file formats
  • Cached research results to avoid re-scraping
  • Configurable validation thresholds
  • Export to PDF/DOCX for sharing

About

End-to-end pipeline that ingests startup pitch decks (PDF/PPT), analyzes company information, performs autonomous web research, and validates startup claims — producing a structured, investor-ready markdown report. Built in Python using OpenRouter APIs (Claude 3.5 Haiku) with orchestrated research, validation, and synthesis agents.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages