AI Content Creation Projects in KalyanAI Content Creation Projects in Kalyan
Projects
People
Cover image for STORM HERD — 30 seconds,
STORM HERD — 30 seconds, one continuous take, no cuts. A herd of white horses stampedes across the slate rooftops of a nineteenth-century city while a thunderstorm tears the sky open behind them. The camera never breaks: it starts at hoof height on a rooftop ridge, lifts into the air as the herd charges past, drops into a chase behind them, threads between two galloping bodies, climbs above the roof valley, then dives and flies backwards ahead of the leaders to the final leap off the roof edge into open air. The whole film is graded to a single cold palette — cyan-steel monochrome, blacks crushed to zero, highlights blown to pure white. The only light source in the entire 30 seconds is the storm. Five lightning strikes punctuate the take, and at the 16-second mark one of them blows the frame completely white for two frames, turning the horses into black cut-outs before the image recovers through a bloom haze. That flash is the hardest part of the whole piece — a full-frame whiteout is exactly where a model wants to cut, and holding the take through it was the technical problem I built the prompt around. Made in CapCut Video Studio with Seedance 2.5. Full prompt and workflow below.
3
4
975
Cover image for Hey Contra fam! 👋 A lot
Hey Contra fam! 👋 A lot of you saw Version 1 of my Interstellar AI project when it won 2nd place in the CapCut Video Studio Challenge recently. The support from this community during that run was absolutely unreal, so I wanted to share the next evolution with you first. Here is Version 2. I wanted to see how far I could push the current AI video stack to mimic high-end cinematic language. I built this entirely from scratch—made in Higgsfield—to try and nail a very specific, gritty, and tactile aesthetic. Dialing in the prompt architecture to get consistent, heavy motion (like the massive wave and the curved space station) took a lot of trial and error, but the level of control Higgsfield offers creators right now is incredible. I’m always refining my prompt workflow for client projects, but passion pieces like this are where the real R&D happens. If anyone is currently building out their own AI video pipelines or wrestling with motion consistency, drop a comment! I’d love to chat about what tools you’re all testing right now.
2
2
962
Cover image for When clients ask what the
When clients ask what the difference is between a generated video and a directed AI video, this is the benchmark. This project explores a high-end, cinematic visual language—specifically targeting a gritty aesthetic. The Workflow: Prompt Architecture: Built initial shot structures using custom Tapebot integration for precise, repeatable visual rules. Motion Generation: Orchestrated complex motion pipelines using Seedance and Nano Banana Pro, focusing on the physical impact of the subjects (like the heavy gallop of the horse) and subtle camera tracking. Sound Design: Grounded the generated visuals using isolated, ambient foley to build tension, proving that premium sound design is non-negotiable for AI content. Available now for AI Video Lead roles and creative motion projects. Let's build something cinematic.
1
908
Cover image for Hey everyone! I recently spent
Hey everyone! I recently spent some time deep in the AI video pipeline, recreating a commercial guide developed by Higgsfield. When orchestrating these types of AI-generated commercials, getting the motion prompts and visual tone perfectly aligned is the biggest hurdle. For this piece, the focus was all on environmental contrast. I had to ensure the prompt architecture supported a smooth transition from a very stylized, pastel morning routine into a rigid, desaturated office space. The fluid tracking shots were crucial here to anchor the viewer in the character's "audio bubble," making the final comedic payoff—where the high-stress environment is completely neutralized by the music—actually land. Would love to hear how others in the community are currently utilizing Higgsfield or similar tools for commercial shot lists and motion direction!
3
594
Cover image for The Foundry — AXIOM ONE
The Foundry — AXIOM ONE (15s spec spot) New spec drop: a cinematic launch film for a product I invented specifically to direct against — AXIOM ONE, a precision-milled titanium earbud. Process notes for anyone building in AI video: One 15s generation, but directed like a real spot: a locked global style prefix (cinematic / Kodak film emulation), a written product spec repeated verbatim, and three engineered cuts inside a single prompt. Continuity across the wider anthology comes from reference stills generated first, then fed into every video generation — the product never drifts. The "hyper-motion" feel is written as in-camera language: high-frame-rate capture, mixed shutter, engineered speed ramps — not "make it fast." This is Film II of four. Same product, four completely different stories, one visual DNA. Open for cinematic AI ad work — launch films, brand spots, product cinema.
1
510