Claude Code in Action cover
← Back to Courses

Claude Code in Action

Project-based

Two days on a real codebase. Explore, plan, code, commit — plus CLAUDE.md, hooks, subagents, MCP and the honest answer on reviewing code you did not write.

2 Days / 14 HoursAI & Generative AILive online with recordings · On-site · Private cohort
HRD Corp Claimable
HRD Corp Claimable
TP No · 10001731699

Program Fee

$599

1-on-1 doubt clearing · WhatsApp group · Post-training Q&A

Request Syllabus / Call Back

Program Overview

Claude Code is an agent, not an autocomplete, and treating it like one is why most developers get poor results. Over two days you work a real codebase: getting Claude oriented before it writes a line, configuring CLAUDE.md and permissions so it cannot do damage, and running multi-file refactors and debugging sessions where you fix a bug in code you did not write. Day two goes further — hooks that enforce standards Claude cannot skip, headless mode in CI, Agent Skills for your team conventions, subagents for parallel work, and MCP to reach your databases and APIs. It closes on the uncomfortable part: how to review AI-written code honestly, what you must never delegate, and whether it is actually making the team faster.

What You'll Learn

  • Run the agentic loop — explore, plan, code, commit — without skipping the first step
  • Configure CLAUDE.md, permission rules and custom slash commands
  • Execute multi-file refactors that do not break three things
  • Debug by reproducing and isolating, not guessing
  • Choose between thinking mode and planning mode, and justify the choice
  • Add hooks that enforce standards the agent cannot skip
  • Run Claude Code headless in CI and on pull requests
  • Build Agent Skills for team conventions and subagents for parallel work
  • Connect Claude Code to your systems through MCP, safely
  • Review AI-written code honestly, and know what never to delegate

Who Can Join

  • Software developers and engineers
  • DevOps and platform engineers
  • Technical leads adopting AI in a team
  • Anyone getting mediocre results from Claude Code

PREREQUISITES

Comfortable in a terminal and with Git

Working knowledge of at least one programming language

Bring your own repository if you can

Curriculum

What Claude Code is

The loop

Installation and first run

Reading a codebase

Lab 0

Total: 14h across 8 modules

Tools & Technologies

Claude CodeCLAUDE.mdAgent SkillssubagentsMCPGitVS CodeJetBrainsCI/CD

Projects Included

  • Lab 0 — Point Claude at a real repo and have it explain the architecture back
  • Lab M01 — A CLAUDE.md for your own repo, plus two custom commands
  • Lab M02 — A real multi-file bug fixed in code you did not write
  • Lab M03 — The same task in both modes, compared on quality, time and cost
  • Lab M04 — A hook that enforces your standards, and a headless run
  • Lab M05 — A team-conventions skill and a test-writing subagent
  • Lab M06 — Claude Code connected to an MCP server on a real task
  • Lab M07 — A shared team configuration and an AI-written PR reviewed

Official Certifications

Course Completion Certificate — issued by CertTulen Academy.
Course Completion Certificate — issued by Anthropic.
Exam — Not applicable.

Assessment is continuous and practical: 8 Applied Learning Labs (60%) and an end-of-course practical assessment (40%). Passing mark 80%, plus completion of all 8 labs and full attendance. The CertTulen Course Completion Certificate is awarded on meeting that standard.

Questions? Contact us for a free counseling session.

Have questions about this course?

Send us your details — we'll respond within 24 hours.

Chat with us