Claude Code for Working Developers
Go from curious to confident. Ship real PRs, author Skills, and use MCPs safely — all with the review discipline that keeps your team's trust.
Who this is for
Target audience
- Engineers who use (or will use) Claude Code daily in a real codebase
- Teams rolling out Claude Code org-wide and need structured ramp-up
- Developers who have tried the tool but aren't getting consistent results
Prerequisites
- Working knowledge of at least one programming language
- Comfort with git, a terminal, and a code editor
- No prior Claude Code experience required
What you'll learn
10 lessons, each built around the same structure: show, tell, do, break it, check. No lesson has more than 15 minutes of passive content before a hands-on moment.
- 1
What Claude Code is and isn't
The three mental models (inline, chat, agent), honest expectations, and the review reflex that separates good users from reckless ones.
- 2
Prompting that holds up
The STAR framework for engineering prompts. Zero-shot vs. few-shot, structured outputs, JSON mode, anti-patterns, and prompt caching.
- 3
Your first hour in Claude Code
Install, auth, first edit, reading the trace, CLAUDE.md done right, undo/revert/reset. Ends with a graded bug fix.
- 4
Real workflows: red/green/refactor
The test-driven loop with an AI partner. Multi-file refactors without catastrophe, plan mode, commit hygiene, recovering from bad steps.
- 5
MCPs in practice
The MCP mental model, filesystem/git/GitHub MCPs, Jira-to-PR workflows, reading production via Grafana and Confluence, per-tool permissions.
- 6
Cost and context discipline
Estimating cost before running a task, diagnosing context bloat, reset vs. continue, cheaper models for cheaper tasks, budget circuit breakers.
- 7
Debugging when Claude goes sideways
Loop breakers, hallucinated APIs, when tests pass but the feature doesn't, the reset recipe.
- 8
Security and safety habits
Never paste secrets, prompt injection from tool output, pre-commit safety hooks, the security-team reflex.
- 9
Authoring your first Skill
Rules vs. procedures vs. checks vs. isolation. Anatomy of a Skill, writing descriptions the matcher loves, sharing across teams.
- 10
Capstone: end-to-end feature delivery
Pick a real task from your backlog and ship it. Green tests, clean diff, PR description, one-paragraph retro. Graded by the lab orchestrator.
What you'll build
Every track includes graded hands-on labs on realistic codebases. No toy examples.
Fix your first bug end-to-end
A realistic codebase with a failing test. Use Claude Code to diagnose, fix, and verify the fix — then write a clean commit message.
Jira-to-PR with a mock MCP
Read a ticket from a mock Jira MCP, implement the feature, write tests, and open a PR. Graded on correctness, test coverage, and diff hygiene.
Implement a feature, tests and all
A 90-minute capstone mini-lab. Implement a small feature from scratch in an unfamiliar repo, with tests, linting, and a clean PR.
Sample lesson preview
Lesson preview
MCPs in practice: the MCP mental model
- What an MCP server actually is — the one diagram that makes it click
- Filesystem, git, and GitHub/ADO MCPs: when each one helps and when it gets in the way
- Danger zones: which tools can write, which tools can read production, and how permissions work
- Hands-on: connect a mock Jira MCP and go from ticket to working code in under 30 minutes
Certified AI Engineer (Claude Code)
Complete this track to earn your CAE badge. Certifications are earned through practical assessment — a written exam plus a hands-on practical — not just quiz scores. Exportable as Open Badges 2.0 and verifiable by URL.
Badges are valid for 18 months, renewable with a short refresh assessment.
Start your team's training
Per-seat annual plans start at $300/user. Enterprise pricing available for teams over 200.
Not sure where to start?
Take our free 3-minute AI maturity assessment and get a personalized recommendation for which tracks fit your team.