AI Meeting Minutes to Action Plans with Notion Sync
Project Overview
Type: Personal Project
Tech Stack: Python, Streamlit, OpenAI GPT, Notion API
Role: End-to-end builder (ideation, development, deployment)
Timeline: 1-week prototype
Summary
AI Meeting Action Plan Generator is a web application that lets users upload meeting transcripts from platforms like Zoom, Google Meet, or Microsoft Teams, and automatically generates clear, structured action plans using OpenAI’s GPT models. It highlights key decisions, task assignments with owners and deadlines, and concise summaries — all in a user-friendly interface. Users can optionally push the generated tasks directly to a Notion database for seamless task tracking.
Problem Statement:
Meeting transcripts are often long and unstructured, making it time-consuming to identify decisions, action items, and responsibilities. Teams waste valuable time manually parsing meeting notes or risk missing critical follow-ups.
Solution:
This tool enables users to:
Upload meeting transcripts (TXT or Markdown) from popular conferencing platforms
Automatically extract:
Action items with assigned owners and deadlines
Key decisions made during the meeting
Summary bullets for quick referencePush these structured action items to a Notion task database with a single click
Built with Streamlit, it offers an accessible interface without requiring technical knowledge.
How It Works:
Upload a meeting transcript file via the web UI
The backend uses OpenAI GPT to analyze the transcript and generate structured outputs
The app displays key decisions, action items, and summary bullets in a clean format
Users can optionally push the extracted tasks directly to Notion for task management
API keys are securely handled via environment variables (.env or Streamlit secrets).
Tools and Techniques
OpenAI GPT: Natural language processing and structured output generation
Streamlit: Interactive frontend UI and app deployment
Python: Backend logic, file handling, API integration
Notion API: Task assignment and database management
Environment Variables: Secure API key configuration
Key Learnings
Designing end-to-end AI-powered applications
Parsing and structuring complex, unstructured text data
Ensuring seamless multi-API integration (OpenAI + Notion)
Managing user experience to maintain context across asynchronous steps
Handling real-world issues like date parsing, error handling, and state management
Real Life Use Cases
Product and project managers summarizing meetings
Customer success teams capturing client discussions
Executive assistants preparing follow-up notes
Teachers and facilitators generating summaries of learning sessions
Government offices automating task tracking from inter-departmental meetings
Try the live demo
Key Outcomes / Results:
Designing end-to-end AI-powered applications
Parsing and structuring complex, unstructured text data
Ensuring seamless multi-API integration (OpenAI + Notion)
Managing user experience to maintain context across asynchronous steps
Handling real-world issues like date parsing, error handling, and state management