i stopped asking AI to write one post at a time.
instead, i built a 24/7 content team inside Grok Bot:
→ monitors X trends
→ hands work between specialist Bots
→ creates content
→ saves what works as Skills + Routines
the full beginner setup ↓
haha no one asked but every letter of this title becomes its own tiny object on hover — an avocado, a sleeping cat, an origami boat. 6 letters, 12 images, no repeats
i'm now UNSTOPPABLE at decorating my own pages thanks to Enter @ConvergeAI_X . try it, it's SO fun
damn, i finally finished my AI short film!!!
made this with seedance 2.5 and holy shit, this is getting INSANE.
i can literally let AI handle the camera direction, long takes, shot transitions, visual storytelling, even the pacing with music. and it all comes together SO FAST. still kinda blows my mind lol.
it's called TUNNEL VISION.
basically, a raccoon paints a fake tunnel to trick a rabbit. the rabbit somehow runs straight through it, and the raccoon just smashes into the wall lmao.
but then he gets obsessed with proving he can make it through too.
the funniest part? HE PAINTED THE DAMN TUNNEL HIMSELF.
kinda stupid, kinda sad, kinda all of us sometimes.🤣
anyway, really happy with how this turned out. hope you enjoy it!!
made with @FramiaPro
rt + comment "TUNNEL VISION" and i'll share my entire AI filmmaking workflow and prompts. (follow so i can DM you)
#AIart️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️ #FramiaCPP #aishorts
damn, i finally finished my AI short film!!!
made this with seedance 2.5 and holy shit, this is getting INSANE.
i can literally let AI handle the camera direction, long takes, shot transitions, visual storytelling, even the pacing with music. and it all comes together SO FAST. still kinda blows my mind lol.
it's called TUNNEL VISION.
basically, a raccoon paints a fake tunnel to trick a rabbit. the rabbit somehow runs straight through it, and the raccoon just smashes into the wall lmao.
but then he gets obsessed with proving he can make it through too.
the funniest part? HE PAINTED THE DAMN TUNNEL HIMSELF.
kinda stupid, kinda sad, kinda all of us sometimes.🤣
anyway, really happy with how this turned out. hope you enjoy it!!
made with @FramiaPro
rt + comment "TUNNEL VISION" and i'll share my entire AI filmmaking workflow and prompts. (follow so i can DM you)
#AIart️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️ #FramiaCPP #aishorts
as a kid my biggest dream was to be buried in burgers.
20 years later, AI...no SI made it happen. LITERALLY raining burgers 🍔
showed this to my nephew and his friends. they watched it twice and said "that's not AI, that's just your weekend."
rude. but fair.
made with @FramiaPro . honestly kids are harder to fool than most adults now… and framia still got close 😂
🤖 Made with AI
as a kid my biggest dream was to be buried in burgers.
20 years later, AI...no SI made it happen. LITERALLY raining burgers 🍔
showed this to my nephew and his friends. they watched it twice and said "that's not AI, that's just your weekend."
rude. but fair.
made with @FramiaPro . honestly kids are harder to fool than most adults now… and framia still got close 😂
🤖 Made with AI
WTF! claude opus 5.5 just made me a whole animated film about love. not a prompt for some video model, it wrote the entire thing in code.
particles for every week of a life, two heartbeats slowly syncing, an actual attention matrix where every word ends up looking at "you"
this isn't AI… no no no. SI. super intelligent filmmaker
love = finite time × undivided attention❤️
WTF! claude opus 5.5 just made me a whole animated film about love. not a prompt for some video model, it wrote the entire thing in code.
particles for every week of a life, two heartbeats slowly syncing, an actual attention matrix where every word ends up looking at "you"
this isn't AI… no no no. SI. super intelligent filmmaker
love = finite time × undivided attention❤️
Opus 5.5 is INSANE !!🫣
i just made a full product launch film for @Meituan_LongCat 2.5-preview… and i didn't open after effects once. not one keyframe by hand.
just me chatting with opus 5.5 → it wrote the whole thing in remotion (react code → video). the halftone xerox look, the camera moves, the transitions… even the MUSIC. it literally synthesized the beat in python and synced every cut to it 🤯
and my favorite part: the cat's eyes 👀 they're the visual anchor of the whole film. they watch you type, blink on the beat, squint, glance at stuff… then land perfectly on the real longcat logo at the end. opus came up with that.
i kept giving notes like "make it less like a powerpoint" and it just… got it. every round got better.
honestly this is the first time i felt it in my gut: AGI is coming.
Opus 5.5 is INSANE !!🫣
i just made a full product launch film for @Meituan_LongCat 2.5-preview… and i didn't open after effects once. not one keyframe by hand.
just me chatting with opus 5.5 → it wrote the whole thing in remotion (react code → video). the halftone xerox look, the camera moves, the transitions… even the MUSIC. it literally synthesized the beat in python and synced every cut to it 🤯
and my favorite part: the cat's eyes 👀 they're the visual anchor of the whole film. they watch you type, blink on the beat, squint, glance at stuff… then land perfectly on the real longcat logo at the end. opus came up with that.
i kept giving notes like "make it less like a powerpoint" and it just… got it. every round got better.
honestly this is the first time i felt it in my gut: AGI is coming.
jev is INSANE !!!!
every time i make an ai film, finding the look means opening way too many pinterest and google tabs.
so i built the tool i wanted with @typesafeai .
one sentence → visual references in as little as 1 second in my tests.
24 directors. 47 films. 282 frames.
all in one visual universe.
type “young love” and shunji iwai’s world starts gathering around your idea.
type “dark satire” and suddenly you’re thinking in wes anderson.
click any frame to explore the composition, lighting, colors and films behind the style.
from “i have a story” to “this is how it could look.”
less tab hopping. more filmmaking.
rt + comment "jev" and i'll send it over.
(follow so i can DM)
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
jev is INSANE !!!!
every time i make an ai film, finding the look means opening way too many pinterest and google tabs.
so i built the tool i wanted with @typesafeai .
one sentence → visual references in as little as 1 second in my tests.
24 directors. 47 films. 282 frames.
all in one visual universe.
type “young love” and shunji iwai’s world starts gathering around your idea.
type “dark satire” and suddenly you’re thinking in wes anderson.
click any frame to explore the composition, lighting, colors and films behind the style.
from “i have a story” to “this is how it could look.”
less tab hopping. more filmmaking.
rt + comment "jev" and i'll send it over.
(follow so i can DM)
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
jev is insane !!! 🤯
i built an ai casting system for ecommerce videos with @typesafeai .
199 ai model profiles. 15 product categories.
661 judgments in ~4 seconds in my tests.
one brief and it:
→ scores all 199 profiles for the campaign
→ narrows them down to 6 finalists
→ ranks 45 products, 8 hooks and 8 scenes
luxury watches? beauty? cars?
change the brief. watch the entire shortlist change.
the wild part is watching hundreds of scores land before i can even scan the first row.
this is the part of making ai ads i’ve wanted to automate forever.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
jev is insane !!! 🤯
i built an ai casting system for ecommerce videos with @typesafeai .
199 ai model profiles. 15 product categories.
661 judgments in ~4 seconds in my tests.
one brief and it:
→ scores all 199 profiles for the campaign
→ narrows them down to 6 finalists
→ ranks 45 products, 8 hooks and 8 scenes
luxury watches? beauty? cars?
change the brief. watch the entire shortlist change.
the wild part is watching hundreds of scores land before i can even scan the first row.
this is the part of making ai ads i’ve wanted to automate forever.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
LLM: yap yap yap 🤓
Jev: pulls lever 💀
🤖 Made with AI
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution





