AI Engineering · ENGLISH · ONE-DAY WORKSHOP

Claude Agent Workshop

Build AI Agents & LLM-Powered Applications with Python

Learn how AI agents work by building them yourself.

Explore the sample outcome View the day’s agenda

15 November 2026

Melbourne · Proposed: 10:00 AM–4:00 PM AEDT
Includes lunch break and two short breaks. Meals are not advertised as included.

Early bird A$328 · Standard A$389

Standard fee: A$389. Early bird: A$328 until 8 November 2026, 11:59 PM AEDT (Melbourne time).

Instructor and exact venue: to be confirmed. Account requirements and any software or API charges will be confirmed before registration.

See the kind of outcome you’ll build

Completed example: document-assisted support agent

Fictional, pre-generated output. This preview does not run an agent. The full workshop reference application is still being prepared.

Order DEMO-1042 is in transit, with estimated delivery on 28 October. The return window is 30 days after delivery, according to POLICY-01. The agent cannot confirm receipt or approve a return.

Evidence and execution

Validate order identifier.
lookup_order returns status=in transit and ETA=28 October.
Retrieve POLICY-01: returns accepted within 30 days after delivery.
Produce the answer using those two sources.
Evaluation: order status, policy attribution and absence of an unsupported approval all pass for this fixture.

About the workshop

This one-day coding workshop introduces agent and LLM application development using Python and the Anthropic Python SDK. Build an agent, connect it to tools and explore how predefined workflows differ from agent-directed execution. Focused exercises also cover evaluations, MCP and retrieval-augmented generation.

The workshop teaches underlying development principles through code. It does not teach ready-made AI productivity tools or AI coding assistants. This is a repeat offering of the October Claude Agent Workshop, covering the same core foundations.

What you’ll learn

1. Building AI Agents: develop an agent from scratch using the Anthropic Python SDK.

2. Tool Use: define tools, validate arguments and return results to the model.

3. Deterministic Workflows: control execution through predefined steps and explicit logic.

4. Non-deterministic Workflows: let the agent select tools and dynamically determine its execution path.

5. Evaluations (Evals): test task completion, tool selection, accuracy and reliability.

6. MCP (Model Context Protocol): explore connections to external tools and systems.

7. RAG (Retrieval-Augmented Generation): retrieve external knowledge to support grounded answers.

Your day, session by session

Five hours of instruction and practical work within a six-hour day. Times are proposed.

10:00–10:30 · Run and Inspect an Agent

Explore the completed teaching example and trace a request through its execution loop.

10:30–11:30 · Build an Agent with Python

Use the Anthropic Python SDK to implement model calls and a basic tool-use loop.

11:30–11:45 · Break

11:45–12:45 · Tools and Workflow Control

Add a tool, handle results and compare deterministic with non-deterministic workflows.

12:45–13:15 · Lunch

13:15–14:30 · Connect Tools and Knowledge

Complete guided exercises with a prepared MCP integration and a small document retrieval pipeline.

14:30–14:45 · Break

14:45–15:40 · Evaluate and Debug

Run a small test set, inspect failures and improve instructions or tool behaviour.

15:40–16:00 · Review and Next Steps

Demonstrate the agent and discuss how to extend it.

What you’ll leave with

A runnable Python agent project.
Examples of deterministic and agent-directed workflows.
Guided MCP and RAG exercises.
A small evaluation set and recorded observations.

MCP and RAG use prepared components. Building a complete MCP server or production retrieval system from scratch is outside this introductory workshop. Attendance does not award an official Claude certification; no exam booking or exam fee is advertised as included.

Who should attend?

Developers, aspiring AI engineers and technical learners comfortable following and writing Python code. Previous experience with Claude Code or Codex is not required.

Before you attend

Bring a laptop with Python, uv and a code editor such as VS Code installed. Prepare an API key capable of making LLM API calls. The exact provider requirement, supported Python version and responsibility for usage charges remain to be confirmed. Claude Code, Codex and other AI coding assistants will not be used; no installation or experience with them is required.

Each workshop is standalone. No software credits, official certification or exam fees are advertised as included.

Other November workshops

8 November 2026 · Business AI