Alta3 Research Logo

Instructor-led training / ai-cce-101

Claude Code for Developers – Beginner

Part of the Claude Code for Developers Series. This full-day course teaches software engineers to master Claude Code fundamentals. Learn to build full-stack applications, review AI-generated code, and understand model selection and cost management.
Request private trainingTalk to us
Instructor-led

ai-cce-101

Duration
1 Days
Format
Live Online
Access
Multiple Timezones

Overview

About this course.

This one-day course gives engineers a working baseline in Claude Code, Anthropic's agentic coding tool. Participants alternate between short lectures and hands-on exercises as they install Claude Code, learn its core workflow, and use it to build a small full-stack application (a backend API and a connected frontend), while developing the habit of reviewing every change before accepting it. The day closes with practical guidance on model choice and cost-aware usage.

Objectives

What you'll learn

  • Install and authenticate Claude Code, and explain its core interaction loop
  • Differentiate conversational mode, Agent mode, and Plan Mode
  • Generate and refine a project-level CLAUDE.md and use file references to scope a request
  • Generate a working backend API from a structured request and use test failures to guide a fix
  • Generate frontend components that match an existing backend contract
  • Use a screenshot to diagnose and resolve a visual bug
  • Choose an appropriate model and monitor session cost

Audience

Who it's for

  • Software Engineer / Software Developer (junior to mid-level)
  • Full-Stack Developer
  • Backend / API Developer
  • Frontend Developer / UI Engineer
  • DevOps / Platform Engineer who writes application or tooling code

Prerequisites

Before you start

  • No prior Claude Code experience required (installation is covered in Module 1)
  • Working familiarity with programming languages (e.g., JavaScript or TypeScript)
  • A laptop with Node.js installed and permission to install command-line tools

Outline

Course outline.

Download PDF

1. Installing Claude Code and the Core Workflow

  • What Claude Code is: a terminal-based agentic coding tool versus inline autocomplete
  • Installing the Claude Code CLI and signing in with the organization's approved account
  • The default Agent loop: plan a small step, edit files, run a command, observe the result
  • Introducing Plan Mode for reviewing an approach before any file changes happen
  • Understanding permission modes: ask, allow, and deny
  • Hands-on: Install, authenticate, and run Claude Code inside a sample repository for the first time
  • Hands-on: Ask Claude Code to explain an unfamiliar file, then make one small, reviewable edit

2. Project Context with CLAUDE.md and File References

  • Generating a baseline CLAUDE.md with /init and reviewing what Claude Code wrote
  • What belongs in CLAUDE.md: build commands, directory layout, and team conventions
  • Using file and folder references to point Claude Code at exactly the right code
  • Configuring permission rules to keep Claude Code out of secrets and build output
  • Hands-on: Run /init on a sample project and refine the generated CLAUDE.md by hand
  • Hands-on: Compare an unscoped prompt against a file-scoped prompt on the same task

3. Building a Backend API with Agent Mode

  • Asking Claude Code for a structural breakdown of a backend before any code is generated
  • Requesting scaffolding in small, reviewable steps rather than one large request
  • Hands-on: Scaffold an Express server entry point, route files, validation, and controller logic
  • Hands-on: Run the local test suite, copy an exact failure into the chat, and ask for a scoped fix
  • Reviewing the generated diff line by line before accepting changes

4. Frontend Integration and Visual Debugging

  • Binding a frontend data model to a backend contract using explicit file references
  • Adding loading, error, and empty states with a single scoped request
  • Hands-on: Build a small dashboard that consumes the API built in Module 3
  • Hands-on: Intentionally break a layout, share a screenshot, and ask Claude Code to diagnose it

5. Cost, Models, and Responsible Usage

  • Choosing between available models for different kinds of tasks
  • Checking session cost and understanding what drives token usage
  • Keeping context focused: starting new sessions for new tasks
  • Closing discussion: treating AI output as a draft, not a finished implementation

For teams

Training a whole team?

Request private training for Claude Code for Developers – Beginner and we'll follow up with a quote for your group.

Request a quoteBrowse the catalog