I build AI automations, dashboards & apps that save small businesses time. Sharing practical AI tools, workflows & what I’m building.
London
Joined August 2010
- Tweets9.1K
- Following1.7K
- Followers899
- Likes5.2K
Kayode fashola retweeted
AI might have made the AI too hard... I've spent days now working on this game and yet to a win. Try a beta build now, free and in-browser!
@threejs @AnthropicAI #threejs #gamedev #indiegame #ScreenshotSaturday
Just built this Grimwood Museum game.
Exhibit 13: a free Halloween anomaly game in your browser. Can you make it to 6 AM?
#indiegame #halloween2026 #halloween
Available to play here: kfplay.itch.io/exhibit-13
Kayode fashola retweeted
Happy #ScreenshotSaturday!
Check out the new teaser for Junkyard Stories: Rebirth!
store.steampowered.com/app/4…
Kayode fashola retweeted
Kayode fashola retweeted
indie studios, hello
how i add fishing to my cozy farm game with Opus 5.5
the trick is adding one new verb per zone. this zone's verb is fishing:
step 1 -> Claude Code drives Unity 6 via unity-mcp, Blender via blender-mcp
step 2 -> village as a new scene, farm HUD carried over as one prefab
step 3 -> Tripo for the set: pier, lighthouse, 4 boats, crab traps, fish stall
step 4 -> Opus rigs the cast in Blender and builds the water with URP shaders
step 5 -> fishing as one system: cast, tension bar, catch card, coins
step 6 -> 12 fish species in a sheet, Opus balances rarity and prices before code
step 7 -> ElevenLabs for splashes, gulls, reel clicks. Cinemachine follows the player
who does what:
> tripo: every new model
> unity 6: scenes, water, lighting
> opus 5.5: c# scripts, rigging, the fish economy
> you: fish until it feels right
one new zone, one new verb. the farm plays the same, the island just got bigger
next zone: the mine or the forest?
game studios are cooked
how i build a cozy farm game solo with Opus 5.5 and Unity 6
the trick is building it like a studio, one department per prompt:
step 1 -> connect Claude Code to Unity 6 and Blender through MCP
step 2 -> write a 1-page doc: plant > water > harvest > sell, 4 seasons
step 3 -> generate every model in Tripo: farmer, barn, windmill, crops
step 4 -> Opus cleans them in Blender and builds the island in Unity
step 5 -> one system per prompt: planting, growth timers, market, coins
step 6 -> Opus playtests with an autopilot and lists bugs before fixing any
who does what:
> tripo: every model
> unity 6: the world, physics, ui
> opus 5.5: scripts, scene setup, playtests
> you: play it and say what feels off
the rule that saves it... build one system, play it, then the next
what should i build next?
Kayode fashola retweeted
BREAKING: Tesla took years to launch the Roadster.
It took 30 mins to add it to the game, and it also does backflips.
Play in your browser: teleoperator.mindblown.ai
Kayode fashola retweeted
20 days in and I still haven't looked at a single line of code.
All made in a single continuous Claude code chat with autocompact set at 250k
No Blender
Just opus 5.5 and @threejs
I built a jet strike game that runs in your browser. No download.
6 jets, a 40 km world, real bombing physics, SAMs you dodge with flares, and a fitted cockpit for every jet.
Runs at 120 fps, even on iPhone and iPad.
Halvard Strike ✈️🔥
Play free: halvard-strike.netlify.app
Desktop, iPad or phone, in landscape or portrait.
Nine years they waited for this motor. Nine years.
And he's still stuck behind the 159 on Oxford Street 😂
Then it gets weird 🚀
🎬 AI parody
#TeslaRoadster
🤖 Made with AI
Kayode fashola retweeted
OMG Tesla new Roadster can fly in San Francisco on.
Literally took one prompt to add the flying car in Teleoperator. Runs on the browser with threejs.
Play here: teleoperator.mindblown.ai/?c…
Super Stock 500 has a new look
The new Super Stock 500 cover puts the racing front and centre: a packed oval, sunset lights, and one more corner to survive.
The latest build also brings smoother high speed steering, a more detailed cockpit, improved crash behaviour, and updated lighting.
Jump into a 40 car race, draft through the pack, and chase the next checkpoint before the clock runs out.
Available on itch.io
Play here for free: kfplay.itch.io/super-stock-5…
🤖 Made with AI
One cab. One shift. The whole City of London at blue hour.
Pick up fares from the kerb, get them across town before their patience runs out, and earn tips for driving like the clock is everything.
Every drop off buys you more time. When the clock hits zero, the shift is over.
Now available to play on itch.io
Play here: kfplay.itch.io/congestion-ch…
Kayode fashola retweeted
this is the game I’ve been working on for the past 10 days.
Purely vibe coded, from the 3D models and animations to the sound, spell effects, and everything else...
It’s basically a 3D arena game for now..
you can fight duels, play 2v2, take on a boss, or play online against your friends or anyone else..
> 10 classes. and 2 more coming..
> every class has its own kit, combos, animations, and sounds
i tested the balance by running thousands of AI fights between every class, keeping win rates between 40% and 60% across matchups..
there’s still a lot to do, but this is where it’s at after 10 days...
this is just a little gameplay showing some of the many matchups you can play..
we’ve got casters, melee fighters, healers...and everything is balanced so that skill makes the difference, and any class can win..
made by
> opus 5.5
> fable 5.1
> astra 6
> sol 6.1
> SWE 2
> K3
Kayode fashola retweeted
$15,000 from one browser game takes 1 to 15 million plays.
I did the math because my first web game, built with Claude Code, went live on CrazyGames today.
Bolt Pit is a Three.js arena shooter. It has earned $0 so far, because CrazyGames keeps ads off during the Basic Launch test, which runs until November 5.
Only games with good numbers get a Full Launch, and that's where the ad revenue share starts.
The formula:
Revenue = sessions × ads per session × eCPM ÷ 1,000 × your share
eCPM is what advertisers pay per 1,000 ads. CrazyGames allows at most one midgame ad every 3 minutes. The share isn't public, and reports range from 40% to 60%, so I used 50%. The rest are my assumptions, based on reported ad rates:
Low: 1 ad per session, $2 eCPM = $1 per 1,000 plays. You need 15 million plays.
Mid: 2 ads, $5 eCPM = $5 per 1,000 plays. You need 3 million plays.
High: 3 ads, $10 eCPM = $15 per 1,000 plays. You need 1 million plays.
One developer reported about €1.20 per 1,000 CrazyGames plays across eight older games. That's the low case, so I plan around the middle at best.
Is 3 million plays realistic?
CrazyGames reported 320 million play sessions a month this October. Poki says it has 100 million players a month. The players exist. Retention and session length decide which line you land on.
That's the bigger reason I'm here. I have Unity games in progress that were meant for the App Store. On the App Store, I'd only find out if people come back after paying for players.
So the new plan is web first, App Store later:
1. Launch on CrazyGames and Poki, where test players are free.
2. Tune retention and session length on real data.
3. If a game becomes a hit on the web, take it to the app stores.
Both portals' exclusivity rules cover only the web, so the app stores stay open.
I'm also skipping CrazyGames' exclusivity bonus. It pays 50% more for two months if your game stays off other web portals, and I want to test on Poki too.
Getting a Claude Code build accepted took six fixes, including a static build, relative paths, no Google Fonts and cutting the size from 20.5 MB to 12.4 MB. One known bug is still live: on landscape phones the start button sits below the fold.
I'll post the real numbers as they come in. Play it here:
crazygames.com/game/bolt-pit
Claude Code Opus 5.5 build this space shooter that uses spatial web audio. Arcade space combat in the rings of a gas giant: launch from the station, fight among the boulders on itch.io.
Headphones high recommended!
Play it here: kfplay.itch.io/halcyon-reach
