Skip to content
← Back to portfolio

AI-assisted SDLC

The software development lifecycle (SDLC) is how a team goes from a requirement to a released change. When AI helps with that work, the lifecycle still needs people who decide, review and own the outcome. This page summarizes how Joe Palala approaches it.

Start from a written spec

Work moves through three plain-text artifacts: requirements, design and tasks. AI can read them, flag gaps and suggest changes. People ask the questions and approve the spec.

Scale review with risk

Formatting and documentation need little or no review. Architecture, security and releases need a human decision. Review effort follows risk, not volume.

Let tests define correct

Unit, integration and end-to-end tests say what correct means, whoever wrote the code. They are the check that does not depend on trusting the author, human or AI.

Keep every change traceable

Every change, from a person or an AI, arrives as a reviewed pull request. Architecture decision records note what was decided and why, so a bad change can be found and reverted.

Keep a person accountable

Every role has a human owner and an AI assistant, and the human is accountable. AI assists the work. It does not own the outcome.

Read the full write-up: Bayanihan Scrum, a collaboration-first workflow.