Claude Academy
Loading

The AI-native SDLC playbook

For engineering, platform, and security leads whose organization already uses Claude Code. Twelve plays for using Claude at each stage of the software development lifecycle (SDLC), so the work around the code runs faster. About an hour of reading, with no quiz or completion badge.

14 lessons1 hr
Start courseSign in to save progress
What you’ll learnBy the end of this course, you’ll be able to
  • Identify what changes at each of the six stages between a traditional and an AI-native SDLC, and choose a first play from the dependency graph
  • Explain how intent.md, a short version-controlled file stating what is wanted, why, and under which constraints, becomes a requirements and design spec in one working session with Claude
  • Choose whether a piece of institutional knowledge belongs in CLAUDE.md, a skill, or a hook
  • Explain how evals rerun whenever the agent's setup changes, how Claude reviews every pull request, and which code people still review by hand
  • Explain how hooks hold actions such as a release until a person approves, and how a production metric leaving its normal range becomes a new intent.md that restarts planning

Many engineering teams still have the same approval gates, reviews, handoffs, and policies, stalling productivity gains made by using agentic coding solutions like Claude Code.

In this guide, we walk through several of our Applied AI team's best practices for integrating Claude internally across each stage of the SDLC to accelerate development and make processes run faster, inspired by working with our customers.

You can also read the full playbook as a single page on the Claude blog.

Who this course is for

This course is for engineering, platform, and security leads whose organization already uses Claude Code and whose approval gates, reviews, and handoffs still run at human speed. It is written for large enterprises, and especially regulated ones, where the review queue cannot be allowed to build and code cannot ship under-reviewed. The plays assume you can change how your organization plans, reviews, tests, and deploys, not only how it writes code.

Prerequisites

To run the plays, you should be comfortable working in Claude Code day to day, and have a Git repository and CI pipeline you are allowed to change. Each play names its own prerequisites under Getting started, and several have none.