https://github.com/virtualpatientengine/eurostempeers2025_workshop
This is a hands-on beginner-level workshop aimed at demystifying AI agents. Participants will learn how to build simple, modular agents using LangGraph, a powerful open-source framework in Python for creating multi-agent workflows
https://github.com/virtualpatientengine/eurostempeers2025_workshop
agentic-ai agents langgraph python3
Last synced: about 1 month ago
JSON representation
This is a hands-on beginner-level workshop aimed at demystifying AI agents. Participants will learn how to build simple, modular agents using LangGraph, a powerful open-source framework in Python for creating multi-agent workflows
- Host: GitHub
- URL: https://github.com/virtualpatientengine/eurostempeers2025_workshop
- Owner: VirtualPatientEngine
- Created: 2025-07-10T15:21:19.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-10T15:40:42.000Z (11 months ago)
- Last Synced: 2025-07-10T22:02:49.738Z (11 months ago)
- Topics: agentic-ai, agents, langgraph, python3
- Language: Jupyter Notebook
- Homepage:
- Size: 2.31 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# EuroSTEMPeers2025_workshop
Welcome to the **EuroSTEMPeers 2025** hands-on beginner-level workshop on **Agentic AI**!
This session aims to **demystify language model agents** by walking participants through how to build **simple, modular agents** using **LangGraph** — an open-source Python framework for creating multi-agent workflows.
## 🧠 LangGraph Agentic AI Workshop
**Hosted by TeamVPE at BioMed X Institute**
Welcome to the hands-on workshop on **Agentic AI with LangGraph**!
This session is hosted by [TeamVPE](https://bmedx.com/research-teams/artificial-intelligence/team-vpe/) at the [BioMed X Institute](https://bmedx.com), proudly supported by [Sanofi](https://www.sanofi.com/en/).
---
## 🎯 Workshop Goals
By the end of this workshop, you'll be able to:
- Understand the **LangGraph** framework for building stateful LLM agents
- Create and manage **agent state** and **tool-calling workflows**
- Use **OpenAI** or **NVIDIA NIMs** as your LLM backend
- Build a **ReAct-style reasoning loop** with tools
- Visualize LangGraph agent flows
- Extend agents with tools for:
- 🧮 Math calculation
- 📊 Matplotlib plotting
- 🌍 Language translation
- 📄 PDF Q\&A pipeline
- 🎨 DALL·E image generation
---
## 🧱 What You'll Build
You will implement several progressively advanced agents:
1. **Echo agent** using LangGraph
2. **LLM-powered assistant** (OpenAI or NVIDIA NIM)
3. **ReAct agent** with intermediate memory and reasoning
4. **Multi-tool agent** with tool routing and control flow
5. **PDF reasoning agent** that can download, extract, and answer questions from papers
6. **Image generation agent** using DALL·E 3
---
## 🧰 Prerequisites
- Python 3.12+
- Jupyter Notebook environment (e.g. Anaconda, VS Code, Colab)
- One of the following API keys:
- [OpenAI API Key](https://platform.openai.com/account/api-keys)
- [NVIDIA NIM API Key](https://build.nvidia.com/)
---
## 🚀 Installation
Install required libraries with:
```bash
pip install langgraph langchain openai matplotlib pymupdf graphviz pydot requests
```
You’ll also need to install:
- `langchain_nvidia_ai_endpoints` (if using NVIDIA NIMs)
- `IPython` (if not included in your environment)
---
## 🔐 API Key Setup
You can either:
- Set environment variables (`OPENAI_API_KEY` or `NVIDIA_API_KEY`)
- Or paste the key directly into the notebook (demo-friendly)
---
## 📚 Topics Covered
| Section | Highlights |
| ---------------- | -------------------------------------- |
| ✅ Basics | LangGraph setup, Echo agent |
| 🧠 LLMs | OpenAI/NVIDIA NIM integration |
| 🔁 ReAct | Reasoning + tool invocation |
| 🔧 Tools | Calculator, plot generator, translator |
| 📄 PDFs | Download, extract, and answer |
| 🎨 Vision | Image generation with DALL·E |
| 📊 Visualization | Flowchart of LangGraph workflows |
---
## 🧑🔬 About the Project
This workshop is part of the **AIAgents4Pharma** initiative — an open-source toolkit for enabling biomedical researchers and pharma professionals to interact with complex data through intelligent, language-powered agents.
👉 Explore more: [VirtualPatientEngine/AIAgents4Pharma](https://github.com/VirtualPatientEngine/AIAgents4Pharma)
---
## 📎 Attribution
Developed by **TeamVPE**, AI research team at [BioMed X Institute](https://bmedx.com). Sponsored by **Sanofi**.
---
## 🧑💻 Get Started
Launch the Jupyter Notebook and start from:
```python
!pip install -q langgraph langchain openai
```
Then follow the steps to build your first intelligent agent!