Agentic SDLC: Ship Software with AI Agent Teams

You're not learning to prompt, you're learning to lead an engineering team that happens to be made of agents. This intensive takes you across the full software lifecycle: turn a messy brain dump into a phased, spec-driven roadmap with OpenSpec, then stand up autonomous workers behind a test pipeline and pre-commit gates where nothing merges until it passes. You'll orchestrate specialist agents working in parallel, coordinating over files and Matrix, with a release-manager agent handling the merges, while you stay the senior engineer, gating everything with checklists, automated security review, and prompt-injection filtering. Give your agents persistent memory that self-corrects from its own failures, kill recurring bugs for good with TDD and Playwright E2E, and run the whole team from your phone. You leave shipping real software, not demos, with control at every checkpoint.

What you leave able to do

• verify a program does what you claimed with a test
• verify model written code against its specification
• design a review gate that catches what the doer cannot see
• reconcile an agents run with a tool that failed
• reconcile what you built with what you can maintain alone
• synthesise an accountability regime for an unattended agent
• synthesise an agent that carries notes across its own runs

How you show it. Paired runs of the same multi-step task with the note store kept and cleared, where the cleared run repeats a step the noted run skips, plus the note text the agent wrote and the later turn that cites it.

It sits on one path

About 40 hours of practice sit behind this one. See what your hours buy.