Claude Code

Claude Code in the Enterprise

Daily Development, Rollout, Security and Governance

Overview

Claude Code is Anthropic's AI coding agent. It works in your terminal, editor or desktop app, understands whole codebases, edits across many files and lets you choose how much autonomy it has. This course follows its journey in a large organisation: first how an individual developer uses it every day, then how a company deploys it securely and at scale.

The first day covers installation and sign-in, navigating and editing multi-file codebases, approval modes, Git, debugging and bug fixing. The second day covers team and enterprise configuration, CI/CD and automation, performance, troubleshooting, security and compliance, and ends with a rollout plan for your own organisation.

Format: Interactive lecture and discussion with plenty of exercises, followed by hands-on implementation in a live-lab environment.

Who it's for

Developers and the people who support them in a larger organisation. It is especially suited to:

  • Software developers looking to accelerate development with agentic AI
  • Engineering leads evaluating AI-assisted coding tools for teams
  • DevOps engineers integrating AI agents into CI/CD workflows
  • Security and compliance staff who review how the tool is deployed

Prerequisites

  • Experience with software development in at least one programming language
  • Basic command-line and terminal usage
  • Familiarity with Git and version control workflows

By the end of this course you will be able to

  • Install and configure Claude Code across local and remote environments
  • Navigate and edit multi-file codebases using natural language commands
  • Use approval modes to balance autonomy with control in enterprise-friendly workflows
  • Integrate Claude Code with Git, CI/CD and team development processes
  • Describe the options for deploying Claude Code in an organisation, with their security and governance controls
  • Draft a rollout plan for your own team or company

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 & orientation 10 min

The course project, the lab environment and a quick round of what your organisation wants from Claude Code.

09:40
Introduction to Claude Code 40 min

What Claude Code is and how it differs from copilots that suggest code in your editor. Its key features, from whole-codebase understanding and multi-file editing to terminal integration, and an overview of the agent loop and how it plans.

Try it: Give Claude Code one multi-file task, watch its plan and list what a copilot could not have done.
10:20
Installation and configuration 55 min

Installing Claude Code with the native installer, Homebrew, WinGet, a Linux package manager or npm (which needs Node.js), signing in with a Claude plan or Console account, or using an API key or a cloud provider. Terminal setup, the VS Code and JetBrains integrations, and managing the context window and token use with /context, /compact and /clear.

Try it: Install and sign in, run claude doctor, connect it to your editor and check the context your first prompt used.
11:15
Break 15 min
11:30
Navigating codebases with natural language 45 min

Asking questions of a codebase instead of searching it by hand, with prompts that follow GCES (Goal, Context, Expectations, Source): using references to files, tracing how modules depend on each other, and restructuring code with prompts.

Try it: Ask Claude Code to explain how a feature flows through five files, then ask it to refactor one part.
12:15
Multi-file editing workflows 50 min

Building a feature that spans several files, keeping dependencies consistent, and testing and validating after every automated edit. Plan mode first, small steps, and checkpoints you can rewind to.

Try it: Add a small feature across three files in plan mode, run the tests and rewind one step to try a different approach.
13:05
Lunch 45 min
13:50
Approval modes and security 45 min

The permission modes from asking before every action to accepting edits, plan mode and the classifier-reviewed auto mode, plus allow and deny rules and sandboxing. Setting boundaries for agent autonomy and reviewing changes before they reach production.

Try it: Set a policy for your team: which actions are allowed, which ask and which are denied, then test it with risky prompts.
14:35
Git integration 45 min

Creating branches and commits with Claude Code, reviewing diffs, resolving merge conflicts with assistance, and generating commit messages and changelogs. Working with the GitHub command-line tool.

Try it: Branch, make a change, review the diff, resolve a planted merge conflict and generate a changelog entry.
15:20
Break 15 min
15:35
Debugging and bug fixing 45 min

Finding the root cause across the codebase from a symptom, applying a targeted patch, running tests, and using the agent for regression testing. Asking for evidence rather than assertions.

Try it: Debug a planted defect from its error message, write a failing test first and confirm the fix.
16:20
Close 10 min

Recap of Day 1 and what we will build on tomorrow.

Outcome: By the end of Day 1 you will have Claude Code installed and configured, and will have edited code, committed and fixed a bug with it under the right approval settings.

09:30
Welcome back 10 min

Questions from Day 1 and the plan for today.

09:40
Team and enterprise features 55 min

Shared context and team configuration through CLAUDE.md, rules and settings checked into repositories, and organisation-wide configuration. Custom rules and style enforcement with hooks. Deployment options: Team or Enterprise plans, the Console, Amazon Bedrock, Google Cloud, Microsoft Foundry, and gateways and proxies. Single sign-on (SSO) and the access controls that come with each option.

Try it: Compare two deployment options against your organisation's requirements and pick one with reasons.
10:35
Security and compliance 45 min

Data handling and privacy: what data Claude Code handles and how it is used under commercial terms, protecting credentials and secrets with deny rules and sandboxing, prompt injection, and audit trails and governance through managed settings, OpenTelemetry usage events and your cloud provider's logs.

Try it: Write the data-handling and secrets rules for your repositories and show how each is enforced.
11:20
Break 15 min
11:35
CI/CD and automation workflows 55 min

Running Claude Code non-interactively (headless mode) with claude -p and the GitHub Actions integration, triggering agent tasks from pipeline events, pre-commit checks through hooks, and automated code review. Keeping automation permissions narrow.

Try it: Build a pipeline step that reviews a pull request with a fixed allowed-tool list and a turn limit.
12:30
Performance optimisation 40 min

Managing large codebases within context limits with nested memory files, path-scoped rules and subagents; splitting work across sessions and worktrees; prompt caching and saved sessions; and choosing the right model tier for the task.

Try it: Reduce the context cost of a long task by splitting it, then compare usage before and after.
13:10
Lunch 45 min
13:55
Troubleshooting 40 min

Diagnosing connection, proxy and sign-in problems, handling agent misinterpretations and bad edits (rewind, clear, restate), and dealing with rate limits and token exhaustion.

Try it: Work through three prepared faults: a proxy error, a misread instruction and a rate-limit message.
14:35
Rollout plan workshop 50 min

Pull the course together into a rollout plan: pilot team, deployment option, configuration, permissions, automation, security review, training and the measures that tell you it is working.

Try it: Draft your rollout plan on one page with owners and a first milestone.
15:25
Break 15 min
15:40
Rollout plan review 40 min

Present your plan to another group, take their questions, and revise it. Review of the key features and workflows from the two days, resources for continued learning and community support.

Try it: Improve your plan with two changes from the feedback.
16:20
Close: next steps 10 min

The step up to configuring Claude Code for engineering teams (subagents, hooks, skills and plugins), and your first actions.

Outcome: By the end of Day 2 you will have a draft rollout plan covering configuration, deployment, automation, security and governance.

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 and a GitHub account. A provided lab environment covers the exercises; if your organisation already deploys Claude Code through a cloud provider, tell us beforehand so we can use your setup.

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 Practice 7 hours (1 day) · 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 £1,50014 hours (2 days × 7 hours)
Enquire