Alex Kubrak

@Osmbproduction Staying motivated for sure. Especially when managing revisions 😅

Honest question for freelance editors: What's the hardest part of your business right now? a) Finding clients b) Pricing your work c) Managing revisions d) Staying motivated

@higgsfield Huge!

Meet Higgsfield CLI + Marketing Skills. Instead of burning tokens on bloated schemas, or shipping broken creative at scale, the CLI keeps agent spend lean and Skills keep output high quality. Pairs with Сodex, Claude Code, Openclaw etc. npx skills add higgsfield-ai/skills

@whoistife_x Feeling lazy? Remind yourself that you’ll feel ten times better after your workout than if you chose to skip it.

If algorithm brings this to you, leave a fitness advice for someone.

@CoachDanGo One of the main investment for sure

People will take you more seriously when you get your body in shape.

@miku_diffract Awesome job man! Good luck in contest

Here’s my entry for #RunwayBigPitchContest I wanted to make a futuristic dystopian series, but not the polished neon version. I wanted grit, fear, real emotion. A future that felt lived in. This is "mei." A series about the most advanced AI model in a synthetic body that can dream, feel, and become something the company refuses to acknowledge: human. When her handler Ube sees the truth, she takes her and runs. Made in @runwayml

@0xbisc @dreamina_ai Sounds good thanks! Should try it

@kyb909 @dreamina_ai It works a bit differently for longer clips You’re building off the previous segment instead of restarting, so it doesn’t drift as much

Into the Swing-Verse made with GPT Image 2 + Seedance 2 #DreaminaCPP @dreamina_ai Prompt below ↓

@letterboxd Fight club and The Devis wears Prada on top is so unexpected

The Top 10 Most Popular Films on Letterboxd This Week 🎤

First paid project done with AI, a commercial spot for a leather goods brand. Seedance 2.0, Nano Banana Pro, Claude for prompts, finished in AE. 5 years of motion design behind me, and until now I only got paid for After Effects work. Now prompts pay too. Not Hollywood, I know. But still - small step, clear direction.

@0xbisc @dreamina_ai Yes sure, but if each new scene needs a fresh storyboard do you still get consistency? Do storyboards + scenes end up looking like the same world, or do they start drifting?

@kyb909 @dreamina_ai Right now it’s limited to 15s per clip, anything longer is stitched together from multiple clips

@maxescu 1. Seedance 2.0 2. AI utomatization 3. Gpt Image

Most fun today with AI: 1. Suno 2. Vibe coding 3. Filmmaking How's it for you?

@pankajstwt woow the first one is just beautiful .. great job!

some recent hero designs from last month

@IamEmily2050 For sure. The scary part isn’t the tech, it’s how fast ‘generated’ becomes ‘remembered’

What we see today in the storyboard is just the beginning; the next few updates will go deeper into each frame in 3D, not just 2D, and people will start making so many fake memories that they believe it is real and want to implement these memories. Perhaps this is how the matrix will start or started, there is no shame in giving yourself the best of what the world can offer.

@ProofofMaro Hours are only half of it. If you grind 10,000 hours learning AI video generation, the credit burn is insane unless you’re funding it with commercial work.”

What have you put 10,000 hours in?

@sama Cheaper/faster is what people feel immediately, smarter wins you over while you’re using it and sticks way harder.

i keep thinking i want the models to be cheaper/faster more than i want them to be smarter but it seems that just being smarter is still the most important thing

@_VVSVS Looks great, love it! Camera move + saturated palette feels exactly like a 00s music video.

Fashion at the Blue Sea

@dilaw006 Looking awesome! Da Nang really is a nice, comfortable city, but lately it's starting to feel more and more crowded. Maybe it's time to check out Nha Trang 👀

as promised, pool reveal. infinity edge over east sea, islands on horizon.. nha trang is a hidden gem nobody talks about enough. da nang gets all the nomad hype but i’m planting my flag here, watch this space.. 🇻🇳

@HBCoop_ Love the workflow but what about the resolution? Packing everything into one grid seems like it would make the details pretty crunchy because of the model's input cap. Does it enough for seedance? Wouldn't separate high-res shots for the actor/world/details yield a better result?

Most people are still prompting video models line-by-line. This is how I’m getting more consistent results with Seedance 2.0 👇 I treat the entire video like a Visual Production Graph — one structured image that acts like a compressed production brief. Instead of describing everything in text, the model can see: • the character (consistent identity, wardrobe, presence) • the world (location, materials, lighting) • the spatial layout (how scenes connect) • the shot sequence (how the story flows) Then I pair that with a short prompt that just handles: → timing → camera movement → shot order The result: Less guessing from the model More continuity across shots Cleaner motion and framing Way fewer regenerations Think of it like giving the model a map + storyboard + lookbook in a single frame — instead of hoping it builds the world from scratch. The seaside villa video below was generated using this exact setup: Visual Production Graph (above) + structured shot prompt (below)