AI-SDLC guides
Last updated
Practical guides for teams evaluating and adopting AI-SDLC. Each guide answers one decision question and links the documentation, specification or white paper behind every claim.
- What is an AI SDLC? Running the software lifecycle with agents, phase by phase
What an AI SDLC (AI-native or agentic SDLC) is, how each of the eight lifecycle phases changes when agents do the work, and where governance, autonomy levels and audit evidence fit.
Last modified
- Why can't teams trust AI-written code yet?
Why review cannot keep up with agent output, and how AI-SDLC moves the human decision to the start of the work and proves the review ran.
Last modified
- What does compliance by design mean?
How AI-SDLC derives gate defaults, attestation and retention from a declared compliance posture, and maps to the EU AI Act, NIST AI RMF and ISO 42001.
Last modified
- How much autonomy does an AI-SDLC agent get?
The four AI-SDLC autonomy levels, how an agent earns promotion, what demotes it, and how cost and readiness gates bound it at every level.
Last modified
- How do you get started with AI-SDLC?
A staged path to adopting AI-SDLC: read the primer, run the getting-started guide on one repository, then decide whether paid help is worth it.
Last modified
- How to evaluate an autonomous SDLC framework
Seven questions, each mapped to a public standard (DORA, NIST SSDF, SLSA, OWASP), with a scoring table for comparing autonomous SDLC frameworks.
Last modified
- Model routing cost: how AI-SDLC bounds LLM spend
How routing cheap models to review probes and strong models to development, plus per-task and monthly budget caps, bounds LLM spend, with a worked end-to-end task cost.
Last modified