Claude Code

Claude Code in Practice

A One-Day Agentic Development Workshop

Overview

Claude Code is not a code-completion tool. It is a development agent that reads your codebase, plans multi-step changes, runs tests and drives a task from one instruction to a working result. This one-day workshop gives you structured practice: configuring it for a real project, writing delegations that produce reviewable results, using CLAUDE.md as project memory, and connecting tools through the Model Context Protocol (MCP).

After a short grounding in what determines code quality and how the agent works, the day is built around three hands-on cases: writing code, reviewing code and designing a component.

Format: A workshop day. Short explanations lead into hands-on work in your own code or a provided project, with the trainer coaching throughout.

Who it's for

Developers who want a fast, practical start with agentic coding. It is especially suited to:

  • Software developers
  • Tech leads and architects
  • Engineers choosing how their team will use Claude Code

Prerequisites

  • Solid experience in at least one programming language
  • Comfort working in a terminal or command line
  • Basic Git knowledge: commits, branches and pull requests

By the end of this course you will be able to

  • Explain the five levers that decide the quality of AI-written code, and use a plan, execute and review cycle
  • Describe how Claude Code reads, plans, acts and verifies, and how it differs from autocomplete and chat tools
  • Configure Claude Code for a real project with CLAUDE.md, skills, hooks, subagents and MCP
  • Delegate an implementation task with context and acceptance criteria, and decide when to accept or redirect
  • Run a structured, traceable code review and a design exploration with Claude Code

Course outline

Times show the standard running order and can be adjusted when we tailor the course to your team. The trainer stays available during breaks to help anyone finish an exercise.

09:30
Welcome & overview 10 min

Objectives for the day, a technical setup check and introductions.

09:40
AI code quality and the plan, execute, review principle 35 min

The five levers behind the quality of what you get: the model, the codebase context, the tooling, prompt precision (we use GCES: Goal, Context, Expectations, Source) and the workflow. Why separating planning, execution and review beats a single prompt, and a decision framework to carry through the day.

Try it: Take one of your recent tasks and say which lever you would pull first, then write the plan prompt for it.
10:15
From autocomplete to agent: the Claude Code paradigm 35 min

How Claude Code differs from editor autocomplete, from assistants such as GitHub Copilot and Cursor, and from chat-based tools, and why that matters. The agentic cycle of read, plan, act and verify, followed by a live demonstration of a multi-file task driven by a single instruction.

Try it: Predict the files a task will touch, watch the demonstration and compare.
10:50
Break 15 min
11:05
Setup and personalisation: CLAUDE.md, skills, hooks, subagents and MCP 90 min

Configure Claude Code for a real project. CLAUDE.md for project memory, architecture rules and team conventions; skills for reusable workflows; hooks for event-driven automation; subagents, which can work in parallel with their own context; and MCP for connecting GitHub, databases and internal APIs, with clear security boundaries.

Try it: Configure your own setup: a CLAUDE.md, one skill, one hook and one connected tool, and test each.
12:35
Lunch 45 min
13:20
Case 1: writing code with Claude Code 50 min

Delegating an implementation task with context and acceptance criteria, then iterating on the result and deciding when to accept it and when to redirect.

Try it: Implement a real feature end to end, with a plan first and a test that proves it works.
14:10
Case 2: code review with Claude Code 50 min

Structuring a review delegation across security, design and performance. A multi-file review that explains what changed, why, and what you should verify.

Try it: Review a real change from the group and write down the three points you would still check yourself.
15:00
Break 15 min
15:15
Case 3: designing with Claude Code 50 min

Generating architecture decisions and Architecture Decision Records from requirements, exploring alternatives and surfacing trade-offs.

Try it: Design a component or feature, compare two alternatives and write an ADR for your choice.
16:05
Recap and Q&A 25 min

Key takeaways, three concrete changes each participant commits to making, and open questions.

Try it: Write your three commitments and share one with the group.

What you'll need

Claude Code

  • A Claude account that includes Claude Code: a Pro, Max, Team or Enterprise plan, or a Claude Console account (the free plan does not include it). Organisations may instead provide access through a cloud provider.
  • A laptop that can run Claude Code (a current macOS, Windows or Linux system with at least 4 GB of memory) and an internet connection. Installation is covered at the start of the course.
  • Git installed. Bring a project you may use for the exercises, or use the provided one.

Practice files are provided for every exercise. You are welcome to use your own non-confidential projects where your organisation's policy allows.

Related courses

Claude Code AI-Assisted Software Development Foundations 14 hours (2 days × 7 hours) · Beginner Claude Code Agentic Development with Claude Code 14 hours (2 days × 7 hours) · Intermediate to Advanced Claude Code Claude Code in the Enterprise 14 hours (2 days × 7 hours) · Intermediate

Not sure which course is right for your team?

Book a free discovery call. We'll talk through your team, your tools and what you want to achieve, then recommend the right course or design a bespoke one.

Book a free discovery call Send an enquiry
From £1,5007 hours (1 day)
Enquire