12 tutorials on delivery. Scope and architect a project, ship features with subagents and TDD, wire CI/CD and security audits, set team governance, then try the Superpowers and Ultraship plugins.
A bootstrapped project repo with CLAUDE.md, rules, hooks and an architecture document
Three features for a task tracker, then a CI/CD pipeline and docs to ship it
A governance design for a 50-person org
A capstone app taken from idea to deployed and monitored
A SPEC.md, ARCHITECTURE.md and PLAN.md for your own side project
Scope a real project, run an AI-assisted research phase, and produce a solid architecture document ready for implementation.
Execute a focused feature sprint using multi-file implementation, TDD flow, and real debugging techniques at production scale.
Automate your pipeline with GitHub Actions, run a security audit, and generate living documentation before launch.
Apply org-wide policies, cost budgeting, audit trails, and permission tiers for AI-assisted development in large organizations.
Configure Claude Code for monorepos using git worktrees, per-package settings, and cross-package workflow patterns.
Define meaningful metrics, track developer velocity, run A/B tests on prompting strategies, and build the leadership case for AI investment.
Apply every Course 2 technique in a single capstone project — from idea to deployed, monitored application.
Install obra/superpowers and observe how the 14-skill plugin hijacks your default workflow — auto-triggering brainstorming, planning, and TDD without a single slash command.
Walk a real feature through the full Superpowers pipeline: brainstorming → writing-plans → subagent-driven-development with two-stage review, ending in a TDD'd, evidence-verified commit.
Install Houseofmvps/ultraship and run /ship — a parallel-tool audit producing the 4-category scorecard (SEO+AI Visibility, Security, Code Quality, Bundle Size) that gates every deploy.
Master the ultraship auditor agents and post-ship lifecycle commands: real penetration testing, GEO/AEO AI-visibility checks, launch copy generation, week-over-week growth tracking, and incident response.
Walk one project — StandupBot, a Slack daily-standup aggregator — end-to-end through all 5 stages: idea → spec → architecture → DAG of parallelizable tasks → worktree-dispatched subagents → ship.
One year of access for $20, or subscribe for $45/month to get this course and every other one.
Not strictly, but it helps. This course assumes you already use the Define → Plan → Implement workflow, CLAUDE.md, hooks and subagents. Those are taught in Claude Code Fundamentals.
Each tutorial lists its own estimate, from 25 to 55 minutes. Across all 12, the listed estimates add up to roughly 7 to 9 hours.
Claude Code, Git and a GitHub account for the CI/CD and Actions work. The Superpowers and Ultraship tutorials install those open-source plugins as part of the lesson.
No. The lessons are about workflow: scoping, sprints, pipelines, governance and monorepos. The monorepo tutorial covers Nx and Turborepo. The Napkin to DAG example is a Slack bot.