AI creative direction for disruptors. no slop
Joined August 2026
- Tweets34
- Following44
- Followers1.4K
- Likes94
Pinned Tweet
i quit my (6fig) investment banking job to go all in on building/creating with AI w/ @beechinour
what i do:
- prompt engineering
- art direction
- ai content systems
i'm brand new to x, so would love to connect with anyone building in content/ai (will be posting more)
i'm dropping a full guide with all the image and video prompts in my telegram channel here: t.me/gt47_lab
i spent the past few weeks working out a repeatable workflow for consistent character sheets that don't drift in animation...
perfecting one ai character is the first hurdle. getting the rest of the cast to match is the real pain point
here are some results from for an upcoming project...
reply “CHARACTER” + RT and i’ll send you the full character sheet prompting guide (must be following so i can dm)
i dropped the most cracked start-frame workflow guide in my tele: t.me/gt47_lab
it covers:
> sourcing style references
> building moodboards in midjourney
> extracting style locks from ref images
> correct prompting language
> refining failed generations
these 4 arcane frames are part of a board i’ve used to create sick images
take these and start testing
GT retweeted
i have built the most innovative AI community in the world...
here's how i used AI agents to build it for me:
- the community lives on Whop, and i control it entirely from their CLI
- every section is its own app, so i push updates to the lessons and ship REAL software inside the community
- my agent has full vision on who's using what and where engagement peaks, so the upsell sits exactly where it should
i firmly believe this is the future of education
GT retweeted
AI video is 100% solved
this guy took it a step further and started an interactive, infinite AI livestream rendered in real time with GPT-6 Astra on Higgsfield
🤝 Paid partnership
This is my last stream as a human being. i'll explain when it's over
first Al streamer in the world and it's me. this stream will never end.
going to bed now. don't make me do anything that can jail me.
powered by GPT-6 Astra and @higgsfield.ai
live on Kick.
Readers added context they thought people might want to know
AI streamers predate this. Neuro-sama has livestreamed as an AI VTuber interacting with chat since December 2022. Real-time infinite AI video livestreams using models like MiniMax H3 Max also launched in late August 2026.
en.wikipedia.org/wiki/Neuro-sama
twitch.tv/vedal987
numerama.com/tech/2321717-c…
kick.com/n3on
GT retweeted
astra in premiere pro is mixing audio 1000x better than i ever could, and i've been editing for 11 years
use this prompt:
----------------------------------------------------------
Act as a professional film sound mixer and fully mix this Premiere Pro sequence.
Use these rules:
Dialogue: always priority. Normal speech around -18 to -12 dBFS, with peaks generally -6 to -3 dBFS.
Dialogue consistency: clip-gain first, then light compression. Avoid obvious pumping.
Music under dialogue: typically -30 to -20 dBFS depending on density. Duck it enough that every word is effortless to understand.
Music without dialogue: can rise to roughly -18 to -10 dBFS.
Ambience/room tone: roughly -35 to -24 dBFS. Never leave unnatural dead silence between edits.
Background Foley: usually -30 to -18 dBFS.
Important Foley/SFX: roughly -18 to -8 dBFS.
Major cinematic impacts, gunshots, explosions, transitions: may peak around -6 to -2 dBFS, but never clip.
Master: never exceed -1 dBTP true peak.
For a web/YouTube delivery, aim around -14 to -16 LUFS integrated while preserving cinematic dynamics.
Sound-design rules:
Dialogue > story-critical SFX > music > ambience.
Use EQ to remove rumble/mud; don't make voices unnaturally thin.
Use subtle compression, automation, fades, crossfades, room tone, reverb and perspective.
Sounds farther from camera should be quieter, darker and wetter; close sounds should be clearer and more present.
Layer impacts with transient + body + low-end/tail when appropriate.
Build sound around cuts, movements, reveals and transitions.
Do not make everything equally loud.
Preserve quiet moments so loud moments actually feel loud.
Avoid clipping, harsh limiting, pumping, abrupt ambience changes or obvious audio cuts.
Continuously A/B the entire scene and make level decisions in context, not clip-by-clip.
Go through the complete timeline, identify every audio element, organize it logically, then perform the actual mix in Premiere. The final result should feel polished, immersive, controlled and cinematic rather than like individually normalized clips.
i’ve been spending 16 hours a day workshopping astra, mj v8.2, and gpt 2 to create undetectable ai anime…
i put together a full breakdown, it includes:
> the best anime moodboards
> extracting color science with ai
> prompt engineering
> my complete workflow
comment “AI” + RT this and i'll send you the full guide (must be following so i can dm)
GT retweeted
how to duplicate the style of any image you find online
here's my 3-step style-cloning system:
step 1 : copy+paste your image inside ChatGPT
step 2 : use this prompt : "extract this visual style as JSON structured data: colors, typography, composition, effects..."
step 3 : save this file and inject it whenever you want to recreate this style
this is stupidly simple but works well for most
i’ve tested every single ai image generation method since i started working on ai video projects...
i think i have finally discovered a few really good frameworks that let me create tasteful, authentic, and high quality start frames for videos
my midjourney workflow and setup is trained off of the coloring/pictures of some of my favorite directors, without stealing them of course
reply “IMAGE” + RT and i’ll send you our full image prompting guide (must be following so i can dm)
