Getting Started with AI-Assisted Development

AI development tools help you write code, create documentation, review work, and automate repetitive tasks — faster and with higher quality. This guide gets you from zero to your first result in under 10 minutes.

What is AI-assisted development?

💬

You describe

Tell AI what you need in plain language: "write unit tests for this service" or "create a sprint planning agenda".

🤖

AI generates

AI produces a draft following your team's standards and best practices — code, tests, documents, or analysis.

You review

You review, refine, and approve the result. AI accelerates your work — you stay in control.

3 steps to get started

1

Install an AI tool

Choose one AI-assisted development tool and install it. Both work with any programming language and any project.

2

Pick your role

Are you a Developer, QA Engineer, Business Analyst, Project Manager, or Solutions Architect? The portal will show content relevant to your role.

3

Try your first skill

A skill is a structured instruction set that guides AI to help you with a specific task — like writing tests, reviewing code, or creating requirements.

Step 1: Install an AI tool

Pick one. You can always switch later. Both integrate with your existing IDE and workflow.

Claude Code

Anthropic's CLI for AI-assisted development. Works in any terminal.

npm install -g @anthropic-ai/claude-code

Skills are installed in .claude/skills/ in your project

Cursor

AI-native code editor built on VS Code. Works with any language.

Download from cursor.com

Skills are installed in .cursor/skills/ in your project

Step 2: Pick your role

The portal organizes content by role. Select yours on the home page to see what's relevant to you.

DeveloperQA EngineerBusiness AnalystProject ManagerSolutions Architect

Step 3: Try your first skill

Here's the best starting skill for each role — chosen for low risk and immediate value:

Important: What AI does NOT replace

  • ⚠️Your judgment. Always read and understand AI-generated code before committing. AI can produce plausible but incorrect output.
  • ⚠️Security review. Never deploy AI-generated authentication, encryption, or access control code without human review.
  • ⚠️Testing. AI-generated tests must actually verify behavior. Check that they fail when the code is broken, not just pass trivially.
  • ⚠️Dependencies. AI may suggest non-existent libraries or outdated versions. Always verify imports and dependency versions.
Browse the catalog →

Or use the AI Advisor on the home page to describe your task