Adoption Roadmap

Adopting AI-DLC is a journey, not a destination. This roadmap provides a phased approach that scales from individual experimentation to organization-wide practice.

Phase 1: Individual Developer

Start with one developer on one project. Master the fundamentals before scaling.

Prerequisites

Before starting Phase 1, ensure you have:

Claude Code installed and working in your development environment
A real project - Not a toy project, but something you're actively working on
Dedicated time - Plan for a learning curve in the first few sessions
An open mindset - AI-DLC may feel slower at first; trust the process

Goals

Complete 3-5 units using the default workflow
Experience all four hats (Researcher, Planner, Builder, Reviewer)
Develop muscle memory for hat transitions
Build intuition for writing good completion criteria

Activities

First session:

Install the AI-DLC plugin
Create your first intent with 2-3 units
Work through the first unit using /elaborate and /construct
Complete the full hat cycle

Subsequent sessions:

Experiment with different workflow types (TDD, Hypothesis)
Practice writing increasingly specific completion criteria
Learn when to /clear and restart vs. when to push through

Success Metrics

Metric	Target	How to Measure
Units completed	3-5	Count `.ai-dlc/unit-*.md` files marked complete
Hat utilization	All 4 used	Self-observation
Criteria specificity	Improving	Fewer ambiguous criteria per unit
Autonomy achieved	Some AHOTL	At least one Builder phase ran without intervention

Common Pitfalls

Pitfall	Symptom	Solution
Skipping research	Jumping straight to building	Force yourself to wear Researcher hat for 5+ minutes
Vague criteria	"Make it work" style criteria	Ask "How would I verify this programmatically?"
Premature optimization	Trying advanced workflows first	Stick to Default workflow until it feels natural
Giving up early	Abandoning mid-unit	Complete at least 3 units before evaluating

Phase 1 Checklist

Phase 2: Team Adoption

Once one developer has proven the methodology, expand to the team.

Prerequisites

Before starting Phase 2, ensure:

Phase 1 champion available - Someone who can answer questions and model good practices
Team buy-in - At least interest, preferably enthusiasm
Low-stakes project - Don't start with the critical release
Documentation ready - Share this documentation with the team

Goals

All team members complete at least one unit independently
Establish team conventions for intent/unit organization
Integrate AI-DLC into existing workflows (PR reviews, standups)
Create shared understanding of when to use which workflow

Rolling Out

Champion Model (Recommended)

Champion demonstrates AI-DLC on a real task in team meeting
Team members pair with champion for their first unit
Team members work independently with champion available for questions
Team reviews first independently-completed units together

Pair Programming Model (Alternative)

Two developers work through units together
One drives, one observes and learns
Switch roles for next unit
Both become champions for remaining team members

Establishing Conventions

The team should agree on:

Convention	Options	Recommendation
Intent file location	Root, `.ai-dlc/`, feature branch	`.ai-dlc/` directory
Unit naming	Sequential, descriptive, both	`unit-01-description.md`
Commit strategy	Per unit, per hat, per feature	Per unit (one commit per complete unit)
PR integration	One PR per intent, per unit	Per intent (all units in one PR)

Integration Points

Daily standups:

"I'm working on unit 3 of the authentication intent"
"Blocked on unit 2 - need API spec clarification"

Pull requests:

Link to intent file in PR description
Reviewers can check completion criteria were met
Unit files provide context for changes

Sprint planning:

Estimate by unit count, not hours
Break features into intents during planning
Dependencies visible in unit DAG

Success Metrics

Metric	Target	How to Measure
Team adoption	100%	All team members completed at least one unit
Consistency	High	Team uses same conventions across projects
Quality	Improved	Fewer bugs, cleaner PRs, better test coverage
Velocity	Maintained or improved	Sprint completion rate

Common Pitfalls

Pitfall	Symptom	Solution
Inconsistent adoption	Only champion uses AI-DLC	Make it required for specific task types
Convention drift	Everyone does it differently	Document and enforce conventions
Overhead complaints	"This slows me down"	Focus on quality improvements, be patient
Tool blame	"AI made a mistake"	Improve criteria and review process

Phase 2 Checklist

Champion identified and prepared
Team introduction completed
All members completed first unit
Conventions documented
Integration points established
Feedback collected and addressed
Ready for broader adoption

Phase 3: Organization-Wide

Scale AI-DLC across the organization with proper support and governance.

Prerequisites

Before starting Phase 3, ensure:

Multiple successful team adoptions - At least 2-3 teams using AI-DLC effectively
Executive sponsorship - Leadership understands and supports the methodology
Training resources - Documentation, examples, champions available
Metrics framework - Ability to measure and report on adoption

Goals

AI-DLC becomes default methodology for appropriate work types
Training program established for new hires and existing staff
Tooling and automation support AI-DLC workflows
Continuous improvement process in place

Institutionalization

Training Program:

Onboarding includes AI-DLC introduction
Advanced workshops for workflow selection and custom workflows
Champion certification program
Regular office hours with experienced practitioners

Tooling Support:

Project templates include .ai-dlc/ structure
CI/CD aware of completion criteria
Dashboards track adoption metrics
Automated reminders for incomplete units

Governance:

Guidelines for when AI-DLC is required vs. optional
Quality standards for intents and units
Review process for custom workflows
Feedback channels for methodology improvements

Cross-Team Coordination

For large features spanning teams:

feature-xyz/
  INTENT.md              # High-level feature intent
  team-backend/
    INTENT.md            # Backend team's intent
    unit-01-api.md
    unit-02-database.md
  team-frontend/
    INTENT.md            # Frontend team's intent
    unit-01-components.md
    unit-02-integration.md
  team-mobile/
    INTENT.md            # Mobile team's intent
    unit-01-screens.md

Each team owns their intent and units, with dependencies across team boundaries explicitly declared.

Success Metrics

Metric	Target	How to Measure
Adoption rate	>80% of eligible projects	Automated scanning for `.ai-dlc/`
Training completion	100% of developers	LMS tracking
Quality improvement	Measurable	Bug rates, PR revision counts, test coverage
Developer satisfaction	High	Surveys, retention

Continuous Improvement

Quarterly reviews - Assess methodology effectiveness
Retrospectives - Collect and act on feedback
Experimentation - Try new workflows, tools, practices
Sharing - Document and spread successful patterns

Phase 3 Checklist

Executive sponsorship secured
Training program launched
Tooling support implemented
Governance framework established
Cross-team coordination working
Metrics dashboard operational
Continuous improvement process active

Timeline Guidance

The timeline for adoption varies by organization size and culture. Here's a general framework:

Phase	Small Team (5-10)	Medium Org (50-200)	Large Org (500+)
Phase 1	1-2 sprints	1-2 sprints	2-4 sprints
Phase 2	2-3 sprints	3-6 sprints	6-12 sprints
Phase 3	N/A	6-12 sprints	12-24 sprints

Key principle: Don't rush. Sustainable adoption takes time. It's better to have one team using AI-DLC excellently than five teams using it poorly.

Next Steps

Developer Guide - Day-to-day usage for individual developers
Tech Lead Guide - Leading team adoption
Manager Guide - Business case and metrics