What is Pydata London 2025

pydata-london-2025 is a Multi-Agent AI Orchestration Workshop that focuses on using the IntelliNode library to create AI workflows for healthcare and wellness scenarios.

Use cases

Use cases include nutrition assessment, multi-model AI applications, and medical predictions based on clinical data, showcasing the versatility of multi-agent systems in real-world applications.

How to use

To use pydata-london-2025, install the IntelliNode library via pip, set up the environment with necessary API keys, and run the MCP server to serve CSV files for AI workflows.

Key features

Key features include multi-agent orchestration, integration of various AI models (text, image, speech), and a medical prediction system utilizing the Model Context Protocol (MCP).

Where to use

pydata-london-2025 can be applied in healthcare, wellness, and educational settings, particularly in scenarios that require AI-driven analysis and decision-making.

IntelliNode Medical Use Cases

IntelliNode is an open-source library for orchestrating AI workflows using graph-based architectures. This repository contains educational examples demonstrating how multi-agent systems can be applied to healthcare and wellness scenarios.

Install Intellinode

# Basic installation
pip install intelli

# With MCP support
pip install "intelli[mcp]"

Environment Setup

Create a .env file in the project root with these keys:

OPENAI_API_KEY=your_openai_api_key_here
ANTHROPIC_API_KEY=your_anthropic_api_key_here

Install project dependencies:

pip install -r requirements.txt

Launch Jupyter:

jupyter lab

MCP Server

To run the MCP server that serves CSV files using Polars:

pip install -r requirements.txt
cd mcp_server
# polars data provider
python eicu_mcp_server_polars.py

(Alternative) Start the server using the Pandas as data provider:

python eicu_mcp_server.py

Lab Overview

Lab 1: Nutrition Assessment with IntelliNode

• OpenAI GPT-4 analyzes client notes, Anthropic Claude creates meal plans.
• Demonstrates connecting multiple AI providers in healthcare workflows.

Lab 2: Multiple Models with IntelliNode

• Showcases text, image, and speech generation in one system.

Lab 3: MCP Medical Prediction with Graph

• Medical prediction system using Model Context Protocol (MCP).
• MCP server serves CSV files using Polars backend.
• Agents predict outcomes from clinical data.

Slides

PyData - Graph Theory for Multi-Agent Integration
https://www.slideshare.net/slideshow/pydata-graph-theory-for-multi-agent-integration/280302074

⚠️ Important Disclaimer

These examples are provided for educational purposes only, and are not intended for actual patient care as presented.

For production deployments, you must implement logging, secure clinical approvals, and establish appropriate governance around the workflow.

Lab Contribution

The use case and examples in this repository were provided by MedWrite.ai.