Work, Revisited — 01: FishingBooker — Line Out I created this project more than a year ago, when ...Work, Revisited — 01: FishingBooker — Line Out I created this project more than a year ago, when ...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
I created this project more than a year ago, when generative AI was still in its infancy—especially for maintaining the same character across multiple scenes.
The concept followed one man from his desk to open water. The problem was that I couldn’t find stock footage showing the same person throughout that journey.
Most of the film was therefore built from carefully selected stock footage and photography. AI was used only for a few very short transitional scenes that were impossible to source. Because the spot moves quickly, these moments only needed to connect the story—not carry it.
The rest was achieved through editing, 3D camera tracking, rotoscoping, UI compositing, kinetic typography and color grading.
Looking back, the strongest decision was not using AI everywhere. It was knowing exactly where it could solve a missing piece while letting traditional motion and post-production techniques do the rest.
This is the first post in my Work, Revisited series, where I return to older projects and examine the decisions, limitations and techniques behind them.
🌊✨ What if following a mermaid led you to a whole new world?
For the CapCut World LA Challenge, I came up with a little fantasy story about a young man who encounters a mysterious mermaid in the ocean and follows her to a hidden island filled with lights, music, and celebration. 🎬
I wanted to make it feel like stepping into a dreamy, dark fairy tale, with beautiful visuals, magical moments, and smooth transitions that make the whole journey feel connected.
This was such a fun concept to explore, and I loved bringing together AI filmmaking, storytelling, and cinematic visuals to bring the idea to life.
A fall beauty ad where Iranian saffron meets Chinese goji 🌙
I didn't start with a product. I started with two ingredients from two cultures, and built a whole brand from them: four concepts, down to one ad.
Made end to end with Arti in Lovart 2.0.
Production-ready, yes. Strategy-ready, not yet. AI can generate the clip, but someone still has to know what the product needs to say.
I don’t lead with how much AI was involved, but I never hide it if asked.
Is AI finally good enough to shoot a product demo on its own, or human touch is still needed?
I generated a 20s product demo with Seedance 2.5 in a single take providing it with one product image reference and an example video transcript (posted in the comments).
Do you disclose...