Prerequisites — Agentic AI Workshop

Required

All participants

A laptop with a terminal

macOS, Linux, or Windows with WSL. You need a working terminal where you can run shell commands. On Windows, install WSL 2 first.

Hands-on

Python 3.10+

Required for all exercises. If you are on macOS, use brew install python@3.12 or install from python.org. On Linux, your package manager likely already has 3.10+.

            # Check your version
python3 --version
          

Hands-on

An API key

We will provide an API key at the workshop — no need to bring your own. If you prefer to use your own key, OpenRouter works well (one key, many models).

Install OpenCode

We use OpenCode throughout the workshop. It is an open-source AI coding agent that works with any LLM provider.

Primary tool

OpenCode — Desktop App (default)

The easiest way to get started. Download from the website.

Download

Alternative

OpenCode — CLI

If you prefer working in the terminal directly.

            # Install
curl -fsSL https://opencode.ai/install | bash

# Verify
opencode --version
          

Strongly recommended

Data work

DuckDB

We use DuckDB for all data exercises. It is lightweight and fast.

            # macOS
brew install duckdb

# Or via pip
pip install duckdb
          

Install guide

Recommended

Bead

A powerful tool for managing dependencies and data in coding projects. We use it in the workshop.

Install

Optional

If you already have it

Git

If git is already on your machine, you can use it to save your progress as we go. We will not cover git commands in the workshop.

Likely pre-installed

Make

GNU Make for running pipelines. Already installed on most macOS and Linux systems.

make --version

Alternative

Claude Code

Anthropic's agentic coding tool. We demo with OpenCode, but Claude Code works if you prefer it.

npm install -g @anthropic-ai/claude-code

Data

We will provide the data We will provide the data — no need to bring your own project. We will work with two public datasets: the Anthropic Economic Index (we provide a pre-cleaned version) and the Census BTOS AI Supplement.mdash; no need to bring your own project. We will work with two public datasets: the Anthropic Economic Index (download pre-cleaned version) and the Census BTOS AI Supplement.