Rokkit200 Academy

The way we build software is changing. Learn to change with it.

Agentic SDLC Playbook

Our first resource. A structured learning journey for teams adopting AI coding agents — covering the mindset shifts, the workflows, the safety rails, and the team-level changes. Not just the tooling.

Built for engineers, tech leads, and engineering managers who want a structured approach rather than trial and error. No prior AI tool experience needed.

A note on context: This playbook was built for Rokkit200’s engineering team. Exercises reference our internal repos and tooling, but the principles, workflows, and safety rails apply to any team adopting AI coding agents. Take what’s useful, adapt what’s specific.

What’s Inside

Phases 1–4 are published. Phases 5–6 are in development and will ship as they’re ready.

[01] Foundations

The business case, the mindset shift from implementation to specification, and getting your tools set up.

[02] Core Skills

The spec-first workflow, context files (CLAUDE.md), hooks, commands, and MCP servers.

[03] Quality & Safety

Code review for AI output, risk tiers, automated pre-filtering, and the five non-negotiable rules.

[04] Fluency & Judgment

Adoption levels from Assisted Coding to Orchestrator, and how team ceremonies evolve.

[05] Orchestration — coming soon

Running parallel agent sessions via worktrees, managing context across streams.

[06] Compounding — coming soon

The metacognitive loop that turns agent output into reusable skills for the whole team.

Start with Chapter 1: Why We're Doing This The business case for AI coding agents and what it means for your day-to-day.