patrickz
Super Cool AI Bro
Run, jump, and explore a platforming world that switches between retro pixels and an AI reimagining. Play online or save it offline.
Play the game ↗PROJECTS THAT BECAME LESSONS
Open the workshop.
Try one piece.
Start with a small working slice of a real project. Most labs use a coding agent; 5 can start with a chat assistant or a spreadsheet. Each one gives you a boundary and a test.
First time building? Begin with a small app ↗Play the games · Labs · Warm-up builds · The stories behind the builds
THE AI GAME MUSEUM
Try the four playable games
Explore a platform adventure, a dungeon quest, a Bangkok comic, or a classic dungeon crawl. Each game has its own page with browser controls and an option to save it for offline play.
Legend of Tokens
Explore an overhead adventure, uncover secrets, and take a sword into a world that spans two visual eras. Play online or save it offline.
Play the game ↗
IDE Zone
Step into a playable comic with Patrick and Su: street fights, Thai phrases, and a Bangkok adventure that keeps turning the page.
Play the game ↗
Castle of the Weights
Choose your adventurer, enter a Norse fantasy dungeon, and find your own way through a turn-based world of monsters, spells, and treasure.
Play the game ↗Open the game collection ↗ · Read how the museum came together ↗
17 project labs
Use public code as a reference, or build the practice version in a new project. Keep sample data fictional. 7 have a live version you can open first.
Game labs (7) · App labs (4) · Website labs (3) · Automation, media and work (3)
Game labs
Plan a game as small playable milestones
Produce an implementation plan with a testable first milestone.
50 min · Some experience · Coding optional
Make a shareable 3D character test page
Build a focused viewer that makes a visual bug reproducible.
60 min · Some experience · Uses a coding agent
Playable versionModel private information in multiplayer
Separate public room state from private player information.
60 min · Some experience · Uses a coding agent
Build a tiny offline roguelike
Finish a one-room game with save, reload and a deliberate offline test.
90 min · Some experience · Uses a coding agent
Playable versionBuild a small language-learning game
Make one playable language encounter with a text fallback.
90 min · Some experience · Uses a coding agent
Playable versionBuild one mobile combat encounter
Make movement, one attack and restart work across keyboard and touch.
90 min · Some experience · Uses a coding agent
Playable versionLearn game fundamentals by building Pong
Finish a small match with deterministic scoring and restart.
90 min · Some experience · Uses a coding agent
App labs
Build a daily quest planner
Create a Today view with a persistent task and a useful evening review.
75 min · Some experience · Uses a coding agent
Prove a phone can reach your API
Retire connectivity uncertainty before adding an AI feature.
75 min · Some experience · Uses a coding agent
Build a tracker that works without the network
Save, export and restore a small set of local records.
90 min · Some experience · Uses a coding agent
Playable versionBuild one secure full-stack interaction
Design and test one create-and-list flow with a server-side user boundary.
90 min · Some experience · Uses a coding agent
Website labs
Build an event page that answers real questions
Create a mobile-friendly event page with one verified booking route.
60 min · Some experience · Coding optional
Build a small Markdown publishing system
Publish two readable posts with a dependable index and safe rendering.
60 min · Some experience · Uses a coding agent
Playable versionBuild a service website people can trust
Create a service page whose claims, scope and contact path are clear.
60 min · Some experience · Coding optional
Automation, media and work
Build a searchable prompt library for your team
Create a small library of reviewed prompts people can search, copy and check against a rubric.
60 min · Some experience · Coding optional
Playable versionBuild a research assistant from sample listings
Normalize and compare listings without silently inventing missing facts.
75 min · Some experience · Uses a coding agent
Organize an AI video production pipeline
Create a file-based shot register with versions, references and review status.
75 min · Some experience · Coding optional
Warm-up builds
Shorter guided builds if you have not coded with AI before.
Build your first small app with an AI coding agent
Create and test one complete local interaction before choosing a production stack.
60 min · Some experience
Prototype a one-level game with AI
Build a small playable loop and use playtesting to control scope.
60 min · Some experience
Build a useful free tool: lessons from Reach
Define and test a small accessible tool with an ordinary-device fallback.
45 min · Some experience
BEHIND THE BUILDS
The stories behind the builds
Build with an AI agent, one proven slice at a time.
Learn to tell an AI coding agent what “done” looks like before it builds, then ship a small app in thin slices you can prove actually work.
When it breaks, debug with evidence, not guesses.
Learn six steps for finding out why something broke (a game, a spreadsheet, a prompt) so the AI works from evidence, not guesses.
I turned a weekend into a Phantasy Star music video.
Space wizards. Expiring credits. A production workflow that became almost as interesting as the film.
I gave three AIs my LinkedIn archive. They built this website.
One hundred and four posts, two years of ideas and zero websites. Here is the exact workflow, the prompts and the guardrails that turned them into patrickz.ai, plus the instructional video I made about it.
Twenty people needed training. I gave my anime pipeline a night shift.
A request to train 20 new team members became a practical test of GPT-6: could the workflow I used for anime music videos make useful training material while I slept?
I had a flight to Thailand. So AI rebuilt Mario with me in it.
In one day, parallel AI agents rebuilt Super Mario Bros. and The Legend of Zelda starring me, upscaled and over the top, playable on my phone. The real lesson was how to direct agents and let a bot judge their work.
Build for one real person first.
Learn to start an AI project from one person's real needs, pick the easiest input that works for them, keep a non-AI fallback, and measure what matters to them.