IN-PERSON PYTHON WORKSHOP
Claude Agent Workshop
Build AI agents & LLM-powered applications with Python.
Learn how AI agents work by building them yourself. Join Roy, AI Engineer at V2.ai, for hands-on Python development with the Anthropic Python SDK, tools, workflows, evaluations, MCP and RAG.
Write Python code yourself. No Claude Code or Codex required.
A complete example, ready to explore.
Fictional order-support assistant
Open the finished response, inspect the tool trace and read the runnable Python source.
See a complete exampleOpen a complete agent example.
Explore a fictional order-support assistant: a finished response, source knowledge, tool results and evaluation checks. This illustrative preview explains the concepts; it is not a confirmed classroom project or a student result.
Pre-generated example · Fictional data · No live LLM or external API calls
Local fixture checks
4 / 4 passed
- PASS · uses tool eta
- PASS · cites retrieved policy
- PASS · unknown order has no fabricated eta
- PASS · unknown order has no sources
These checks validate this small example only. They do not establish real-world model accuracy or production reliability.
Understand it. Code it. Evaluate it.
The day focuses on development principles and hands-on Python coding. Topic order below follows the workshop outline; individual session times are not yet published.
01Building AI Agents
Develop an AI agent from scratch using the Anthropic Python SDK. Understand how its core components work together.
02Tool Use
Enable your agent to call external tools and APIs to carry out tasks beyond generating text.
03Deterministic Workflows
Build AI workflows with predefined execution steps and explicit logic.
04Non-deterministic Workflows
Build agentic workflows that make decisions and dynamically select an execution path.
05Evaluations (Evals)
Test and evaluate agent performance, accuracy and reliability.
06MCP (Model Context Protocol)
Connect AI agents to external tools and systems through MCP.
07RAG (Retrieval-Augmented Generation)
Retrieve external knowledge and provide it to an LLM to support more accurate, context-informed answers.
What you will practise
Practise building an AI agent, connecting tools and APIs, defining fixed and dynamic workflows, evaluating behaviour, and working with MCP and external knowledge.
Who this is for
Developers, aspiring AI engineers and technical learners who want to build AI applications with code. Basic familiarity with Python will help you follow the hands-on exercises.
Content provided by Anthropic. Based on Claude Certified Architect – Foundations. Attendance does not award an official certification.
Set up your laptop before Sunday.
Bring your laptop and prepare these four essentials before attending.
LLM API key
Prepare a key that can make LLM API calls. Confirm the supported provider and any usage charges with JR Academy.
A code editor
Install an IDE or code editor, such as VS Code.
Python
Install Python. Confirm the recommended version with JR Academy.
uv
Install uv to manage Python projects and dependencies.
Bring your laptop with Python, uv and a code editor such as VS Code installed. Prepare an API key that can make LLM API calls. Claude Code, Codex and other AI coding tools will not be used; no installation or experience with them is required.
Engineering experience, brought into the classroom.
Roy Luo
Forward Deployed Engineer
Forward Deployed Engineer, focused on Applied AI and AI Engineering, with a strong background in Software Engineering
- Claude Certified Architect – Professional
- Claude Certified Architect – Foundations
- Claude Certified Developer – Foundations
Before you register
Will we use Claude Code or Codex?
No. You will write Python code directly. No AI coding assistant installation or experience is required.
What should I prepare?
Bring your laptop with Python, uv and a code editor such as VS Code installed. Prepare an API key that can make LLM API calls. Claude Code, Codex and other AI coding tools will not be used; no installation or experience with them is required.
Which API key and Python version do I need?
Please contact JR Academy for the supported API provider, recommended Python version and any API usage charges before attending.
Where is the workshop?
Melbourne CBD. The exact venue will be announced. The session runs from 10:00 AM to 4:00 PM AEDT on Sunday, 25 October 2026.
Does attendance award an official certification?
Attendance does not award an official Claude certification. An exam booking or exam fee is not advertised as included.
Join Roy in Melbourne.
Sunday, 25 October 2026
10:00 AM–4:00 PM AEDT
Exact venue to be announced
Workshop fee
$389
Registration uses JR Academy’s existing course application flow. No payment is made by opening the form.