Phase 1
Ideation
The agent as product initiator, not a spec-taker
Read the phase →Course
AI-Driven Development Lifecycle
ADLC (AI-Driven Development Lifecycle) is an approach where AI agents — not people — run the entire software development lifecycle: from the product idea to running it in production. Other names for the same thing: AI-assisted SDLC, agentic SDLC, building software with AI agents. If you automate software creation with AI — Claude Code, Cursor, agentic pipelines — and keep hitting the question «what does a human even still do at this step», this course is about your problem.
Here the development lifecycle (SDLC) is handed to AI phase by phase. At every step there is a provocative limit: what becomes of the phase if there is no human executor in it at all. Not a faster engineer, but a handed-over role: the agent invents the product, talks to stakeholders, plans, assembles the team, writes and reviews code, ships to prod, holds the pager, and closes the loop. An honest boundary in every phase.
The course runs as a ring: 14 SDLC phases where the output of the last (feedback) returns to the input of the first (the idea) — and the cycle turns on its own. The dashed 13→1 arrow is the closing of the loop. Start with the manifesto.
Phase 2
Discovery
The agent talks to stakeholders itself
Read the phase →Phase 3
Marketing
The agent plans the go-to-market
Read the phase →Phase 4
Requirements
The agent turns pain into a backlog
Read the phase →Phase 5
Architecture
The agent-architect and ADRs
Read the phase →Phase 6
Formation
The PM-agent assembles the agent team
Read the phase →Phase 7
Planning
The PM-agent slices and assigns the work
Read the phase →Phase 8
Implementation
The agent writes the code
Read the phase →Phase 9
Code review
Critic-agent versus author-agent
Read the phase →Phase 10
Testing
The QA-agent owns strategy and the run
Read the phase →Phase 11
Release
The agent ships to prod
Read the phase →Phase 12
Operations
The agent on call
Read the phase →Phase 13
Maintenance
The agent closes the feedback loop
Read the phase →Phase 14
Governance
Who holds the loop: the limit of ADLC
Read the phase →Building AI-driven delivery in-house?
Designing the ADLC loop: where the agent holds the role and where the human principal remains — for your team and product.
Email meThe transition engine
Next Move Engine — the system that takes a team to an autonomous delivery loop.
Next Move Engine →