Inspiration
I have worked many years in the film and tv production industry, before becoming a software engineer. Somehow I wanted to put my passion for film and tech together in a project. Inspired by legendary film industry tools like Final Draft (scriptwriting) and Final Cut Pro (video editing), I set out to build their AI-assisted evolution. I created Lucid Lens to democratize storytelling, empowering anyone with a vision to step into the role of a showrunner.
Problems it solves
Filmmaking is traditionally gated by high costs and technical complexity. Independent filmmakers often spend years chasing the funding required to produce a single project. Lucid Lens offers these creators a way to get their project started.
While the demand for online video is exploding, the production process remains grueling and time-consuming. The friction of moving from idea to script, and script to video, often stifles creativity. Lucid Lens makes that journey easy—and fun.
How it works
Lucid Lens is an end-to-end production studio:
AI-Assisted Scriptwriting: A professional editor supporting the Fountain Markdown format, integrated with a real-time, Gemini-powered creative assistant. No professional writing experience? No problem.
Media Gen: Converts video scripts into cinematic storyboards or final video clips. Instantly visualize a series pilot or generate all the assets needed for your next YouTube video.
Notes to Scripts: Scans handwritten notes via your mobile device and transforms them into structured scripts using OCR. Move from a rough idea to a full-blown screenplay with zero manual data entry.
How I built it
The app is built entirely on the Google Gemini AI ecosystem, utilizing the Google Gen AI API, React, Bun, Elysia.js, SQLite, and Inngest.js (for durable executions).
The Brain: Gemini’s long-context window manages entire scripts to ensure narrative consistency.
Media Generation: We utilized Gemini Nano Banana, and Gemini Veo to allow creators to transform scripts into images, storyboards, and high-fidelity video. We made extensive use of advanced features like video extension, image referencing, and frame-specific prompting.
Natural Workflow: We utilize Gemini’s vision capabilities for OCR and its generative power to help creators transform simple ideas on a napkin into a full blown film script.
Challenges I ran into
OCR Accuracy: Perfecting handwriting recognition remains a challenge, particularly for difficult-to-read notes.
Script Editing Software: Building a robust text editor from scratch is difficult. Handling complex formatting and precise cursor movements across different browsers required significant fine-tuning.
Time constraint: This app was built in just over a week. While I leaned heavily on LLM coding assistants (including Google Antigravity), the short timeframe required intense work.
Accomplishments that we're proud of
- Successfully bridged professional scriptwriting standards (Fountain Markdown) with generative AI, allowing Gemini to edit scenes in the exact format industry pros expect.
- Created a seamless workflow where a user can move from a coffee-shop note to a rendered scene in under five minutes.
- Ready for Prime Time: Delivered a fully functional, almost production-ready application in record time.
Next Steps
- AI-Assisted Video Editor: Expanding the suite to include post-production tools.
- Character Generation Tab: Helping creators maintain visual consistency for their cast throughout the film.
- "The Wall of Dreams": Launching a Google-sponsored competition where creators use Lucid Lens to turn their actual dreams into videos, featured in a public digital gallery.
Log in or sign up for Devpost to join the conversation.