After doing these courses, I feel confident creating professional visualizations and dashboards
Shoebur RahmanProcurement Specialist, Giz Bangladesh
Course Description
Why Claude Code for Real Software Development?
Claude Code brings AI assistance directly into your terminal — but getting reliable results takes more than just asking it to write code. You'll learn the structured workflows that separate productive AI-assisted development from frustrating trial and error, starting with how to explore an unfamiliar codebase and build a mental model fast.How Do I Get Better Results, Consistently?
The explore-plan-code workflow is the foundation. You'll use plan mode and extended thinking to separate exploration from execution, and create CLAUDE.md files that give Claude persistent context about your project so you're not re-explaining the same things every session.How Do I Make Claude Code Work the Way I Do?
Claude Code is general-purpose by default. You'll extend it with custom agents for specialized tasks like code review and security auditing, reusable skills that encode repeatable procedures, and hooks that automatically run tests and linting without you having to ask.How Do I Connect Claude Code to Real Data?
Claude can't query your database without a bridge. You'll configure MCP servers to give Claude direct access to a SQLite database, then use that connection to explore schemas, build validated API endpoints, and write tests against real data — all from the terminal.How Do I Ship AI-Generated Code I Can Trust?
You'll bring everything together by planning and building a complete CLI tool, then applying a seven-item checklist to systematically catch the mistakes AI commonly makes before they reach production. Real workflows, real code, real results.Feels like what you want to learn?
Start Course for FreeWhat you'll learn
- Use Claude Code's CLI and explore-plan-code workflow to systematically navigate unfamiliar codebases and implement features without jumping straight to code generation.
- Create CLAUDE.md files, custom agents, and reusable skills that give Claude persistent project context and encode repeatable team workflows across sessions.
- Explain how PreToolUse and PostToolUse hooks work in Claude Code, and apply shell and Python scripts to automate notifications and logging in a development workflow.
- Connect Claude Code to a SQLite database using an MCP server to explore schemas, build validated Flask API endpoints, and write integration tests against real data.
- Apply a code review workflow using a custom agent to systematically identify common AI mistakes — including hallucinated imports, missing error handling, and unvalidated inputs — before shipping.
Prerequisites
Curriculum
Course outline
1
Getting Started with Claude Code
Learn what Claude Code is, how it differs from other AI tools, and see it in action explaining code and finding bugs.
- See Claude Code in Action50 XP
- Getting Started with Claude Code100 XP
- Debug Python Code with Claude Code100 XP
- The Explore-Plan-Code Workflow50 XP
- Initialize and Explore a Codebase100 XP
- Deep Analysis with Extended Thinking100 XP
- Creating Effective CLAUDE.md Files50 XP
- What Belongs in CLAUDE.md?100 XP
- Keeping CLAUDE.md Current100 XP
2
Extending Claude Code
Extend Claude Code with custom agents and skills, then apply them to refactor legacy code with confidence and automate checks with hooks.
3
Connecting to External Tools with MCP
Extend Claude Code by connecting it to a database. Build one working endpoint from exploration to testing.
4
Building a Music CLI
Build a complete CLI tool from scratch, combining every major skill from the course — plan mode, MCP database connections, custom agents, and code review.
S
Software Development with Claude Code
Course
Complete
