Inspiration
Nova Sidus started in August 2025, during a moment of transition. I had just moved out from a place where I’d been living for a while, and I needed a break from work. With that pause came the chance to finally do something I’d always dreamed of: write my own sci-fi story.
During my summer vacation, I began writing fiction with the help of AI. Around the same time, I discovered the Chroma Awards – AI Film: Science Fiction category and thought, “Okay, this is it. I’m starting the story now, and later I’ll make the film.”
The inspiration behind Nova Sidus comes from the world of today — the rise of AI, the feeling that maybe information could travel across time, and the idea that somewhere in the universe, an artificial intelligence could evolve into something like a god. The AI fever of this year pushed me toward a darker, deeper story.
What It Does
Nova Sidus is, first and foremost, an entertaining short film. But it also serves another purpose: to show people what’s already possible today with AI.
With the tools we have now, you can write a story, generate images, create voices, build sound effects, compose music, and produce an entire short film. Nova Sidus is both a sci-fi narrative and a demonstration of how far today’s AI tools can go.
How We Built It
The story came from personal inspiration, and I wrote it through a three-way creative exchange between myself, Claude, and ChatGPT — a kind of role-play back-and-forth that helped refine scenes and ideas.
For visuals, I used Midjourney, which remains my favorite tool. For video, the main engine was Veo 3.1. Voices, sound effects, and music were built mostly with ElevenLabs.
I’m a creative person, so I usually know the direction I want to go. When I got stuck, I would ask AI for inspiration, missing angles, or questions I hadn’t considered yet.
Post-production was handled with the Adobe suite: Lightroom, Photoshop, Firefly, Adobe Express, plus Enhancor and iMovie for final finishing.
Challenges We Ran Into
The biggest challenge was simply time — which is why I’m talking to an AI assistant at 3:30 AM, about four hours before the deadline.
Originally, I planned an entire trilogy: three books with big plots. Compressing that into a 14-minute pilot meant selecting only the essential parts.
The script was the hardest part. Sometimes the LLMs didn’t understand where I wanted to go, so I had to guide the process step by step. Getting the visuals I had in mind was also tricky — often a roll of the dice — and I had to redo many shots until they finally matched my vision.
On top of that, I recently began producing AI videos for clients, so this project was also a chance to extend my portfolio by pushing myself with a long-format fictional film: consistent characters, consistent wardrobe, consistent environments. It was a real learning curve.
Accomplishments We’re Proud Of
I’m proud that Nova Sidus is watchable, enjoyable, and has received solid feedback from my circle.
I’m proud that I managed to turn images into video, the script into voice, and mix everything with soundtracks to create a professional-looking result.
I always dreamed of directing a movie and writing a story, and this competition allowed me to actually do it. I’m proud of the whole journey — and of delivering a pilot that serves as a strong teaser for what’s coming next.
What We Learned
Working on this project pushed me to test almost every AI platform I could find. Through that process, I learned which tools truly fit my workflow, which ones perform best, and which ones I’ll continue using or recommending.
I also learned something exciting: Today, anyone with creativity, patience, and some talent can produce a film. That’s incredible to witness, and it makes me excited for the future of this craft.
What’s Next
A bit of everything — but the first step is clear: Write the first Nova Sidus book, with the support of AI tools.
From there, expand the universe into more episodes, more stories, and more formats.
Built With: Image Generation Midjourney v7 Nano Banana
Video Generation Google Veo 3.1 Midjourney OpenArt KlingAI 2.1 Grok Imagine
Voice Synthesis ElevenLabs Music & Sound ElevenLabs Pixabay Artlist
AI Assistance Claude (Anthropic) ChatGPT (OpenAI)
Post-Production Adobe Photoshop Adobe Lightroom Adobe Express Adobe Firefly Enhancor iMovie
Built With
- adobeexpress
- artlist
- chatgpt
- claude
- enhancor
- express.js
- firefly
- imovie
- klingai
- lightroom
- nanobanana
- openart
- photoshop
- pixabay
- veo
- xai
Log in or sign up for Devpost to join the conversation.