AI-DLC 2.0Workflows, explained

Home / Lesson 3 of 6

How It's Built: Structure

The big ideaAI-DLC 2.0 is five reusable libraries and tools, and every stage runs the same loop: clarify, plan, do, check, retry, and ask for help if stuck.

Video, 2:52. On-screen captions are built in.

The structure follows three habits good engineers already have: DRY (don’t repeat yourself), separation of concerns, and using each tool’s best native features rather than the lowest common denominator.

1. Persona-Agents

A library of specialist agents, like a real team: an architect, a tester, a security reviewer. Each brings focused expertise to a shared goal.

2. Stages

A library of stage definitions. Each says what work must be done and which artefacts must exist when it is finished.

3. Knowledge

A library of reference material the agents use to generate their outputs and to check themselves.

4. Orchestrator

Uses AI to compose stages into an adaptive workflow, then tracks and manages the run. (Lesson 4.)

5. Package manager

Turns the above into tool-specific artefacts, for example for Kiro or Claude Code. (Lesson 5.)

Try it: the self-correction loop

Pick a scenario and step through what the AI does. This is the flow from Section 3 of the spec.

YesNo NoYes self-correction loop NoYes Start Need clarification? Unclear? Consult a human Create / update plan Execute plan Self-verify Pass? Limit hit? Done
Step 0Attempts 0 / 3Human asks 0

Press Next step to begin.

The limit of 3 attempts is just for this demo. In real use the halting condition is a maximum number of iterations or a token budget.

Two kinds of checks

Compartment 2 (the checks) can hold two very different kinds of rule. Knowing which is which is the key to safe autonomy.

Inferential

Written as LLM instructions

For judgement calls that are not a strict pass/fail, such as most code review guidance (“functions should be easy to read”).

Catch: when the same model writes and grades the work, it can satisfy the letter of a check but not its intent. So a stage with only these checks never approves itself; a human still validates the output.

Computational

Written as executables

Scripts, linters, tests, static analysis. For rules with zero tolerance, such as “no code path deletes a CloudFormation stack in production” or “no API endpoint is exposed without authentication”.

Payoff: the more of your rules live here, the more stages can finish (self-halt) without a person.

Try this at work

For each rule your team cares about, ask: is it a judgement call (inferential) or can a script decide it (computational)? Move the scriptable ones first.

Want the original wording? Read Section 3 of the specification.

Quick check

Three questions. Get them all right to mark this lesson done.

Q1. A stage has only LLM-judged (inferential) checks. What happens when its output passes them?

Q2. “No API endpoint shall be exposed without authentication.” How should this be verified?

Q3. In the self-correction loop, what happens when verification fails and the halting condition has NOT been reached?