One concept. From the actual docs.

claude code

5 episodes · code.claude.com/docs · self-hosted clips (also on LinkedIn)

Ep 1
Your first session

Launch claude in any repo, ask in plain English, and it reads your code to answer — read-only, nothing touched.

Ep 2
The agentic loop

Describe a task; Claude reads, proposes a diff you approve, then runs the tests. Read, edit, verify.

Ep 3
CLAUDE.md memory

One file your agent reads every session. Run /init to draft it from your codebase — stop re-explaining your project.

Ep 4
Plan mode

Claude shows its plan before touching a file. Read it, edit it, approve — then it runs. No surprise edits.

Ep 5
Slash commands

Built-in / commands, plus your own saved in .claude/commands/ and shared with your team via git.

the rig

How Dashecorp's autonomous engineering pipeline reviews code.

The Rig · Dashecorp
How The Rig reviews code — meet Review-E

When an AI agent opens a pull request, another agent reviews it. The full review flow in The Rig — Review-E's persona, and the four gates before a change can merge. ~44 seconds.