LLM AGENTS WORKSHOP · ENVIRONMENT SETUP

Initialize Your Agent Workbench

Configure your local Python environment, model runtimes, and LLM frameworks before Day 1. Everything you need to build autonomous systems.

Start 4-Step Setup Workshop Overview

Python 3.11+·Ollama Local Runner·LangChain & LlamaIndex

Core Agent Ecosystem · Live Tools12 Tools Ready

OpenAI AgentsFunction Calling & Tools

Claude 3.7Extended Thinking & MCP

LangChainChains & ReAct Loops

Ollama RunnerLocal Offline LLMs

Google GeminiMultimodal Reasoning

LlamaIndexRAG & Vector Retrieval

MCP ProtocolAnthropic Standards

Hugging FaceOpen Model Hub

Meta Llama 3Open Weights Engine

Mistral AIEfficient Function Calling

DeepSeek R1Reasoning Runtimes

Python 3.11Core Runtime Engine

Downloads & Links

Python 3.11+

python.org/downloads

Ollama

ollama.com

LangChain

pip install langchain

LlamaIndex

pip install llama-index

Workshop Notebook

Shared on Day 1

Setup Guide · LLM Agents Workshop

Get your machine
ready in 4 steps.

Everything you need for the hands-on sessions — Python, a local model runner, and the agent frameworks. Follow the guide below before Day 1.

Hover a line to learn more

install.sh

1# Step 1 — Install Python 3.11 (if not already)

2# Download from https://python.org/downloads

4python --version

5# Expected: Python 3.11.x or 3.12.x

7# Step 2 — Create a virtual environment

8python -m venv llm-env

9source llm-env/bin/activate # Windows: llm-env\Scripts\activate

11# Step 3 — Install core packages

12pip install langchain langchain-community

13pip install llama-index openai

Four things to install. That's it.

The entire workshop toolchain fits in a single virtual environment. Below is every package you'll need, why it matters, and how to verify it's working before Day 1.

Python 3.11

The entire workshop runs on Python. Install 3.11 (or 3.12) from python.org. During install on Windows, tick **"Add Python to PATH"** — this is critical.

Verify Install

$ python --version

Python 3.11.9

$ pip --version

pip 24.0 from ...

Ollama (Local LLM)

Ollama lets you run Llama 3, Mistral, and Gemma locally — no API key needed. Download from **ollama.com**, install, then pull the workshop model.

Pull Model

$ ollama pull llama3.2

pulling manifest ...

success

$ ollama run llama3.2

LangChain + LlamaIndex

These are the two primary agent frameworks used in the workshop. LangChain powers tool-use and ReAct agents; LlamaIndex handles RAG pipelines and data queries.

Quick Check

import langchain

print(langchain.__version__)

0.3.x

import llama_index

VS Code + Extensions

VS Code is the recommended editor. Install the **Python** extension by Microsoft, and optionally **Jupyter** for interactive notebooks. The Ruff linter extension is also helpful.

Extensions

ms-python.python

ms-toolsai.jupyter

charliermarsh.ruff

GitHub.copilot

All commands, ready to paste.

Run these in order in your terminal. Each line is individually copyable.

Create virtual environment

python -m venv llm-env

Activate (macOS / Linux)

source llm-env/bin/activate

Activate (Windows)

.\llm-env\Scripts\activate

Install LangChain

pip install langchain langchain-community langchain-ollama

Install LlamaIndex

pip install llama-index llama-index-llms-ollama

Install OpenAI SDK (optional)

pip install openai

Pull Llama 3.2 via Ollama

ollama pull llama3.2

Run quick sanity check

python -c "import langchain; print(langchain.__version__)"

Need help? Bring your laptop to the venue at least 20 minutes early. Facilitators will be on hand to help with any installation issues before the first session begins.