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
Back to feedPost
Café Joyeux — AI-Assisted Promotional Video Production
Project Type: Brand Promotional Video / AI Video Production / Video Editing
Role: AI Video Creator, Creative Planner, Prompt Engineer & Video Editor
Tools Used: ChatGPT, Google Flow, Adobe Premiere Pro
Project Overview
I developed an AI-assisted promotional video concept for Café Joyeux, an inclusive café and coffee brand built around quality coffee, human connection, and workplace inclusion.
The objective was to create a visually engaging advertisement that highlights the brand's distinctive yellow-and-black visual identity, original mascot-style illustrations, coffee preparation, branded cups, coffee products, and welcoming café atmosphere.
Following the client's feedback to make the advertisement more brand-focused and playful, I worked on a creative direction that combines realistic café scenes with hand-drawn illustrations, coffee-bean doodles, heart accents, and product-focused animation.
The visual approach was designed to feel warm, authentic, recognisable, and consistent with Café Joyeux's identity.
My Role & Responsibilities
1. Creative Planning & Storyboarding
Developed the advertising concept and visual direction.
Planned the sequence of scenes, from brand introduction and coffee preparation to customer experience, product presentation, and the final logo animation.
Structured the advertisement into individual video clips, each generated with an 8-second duration.
Planned transitions and visual continuity between scenes.
2. AI Prompt Engineering
Used ChatGPT to develop the creative concept, prepare the storyboard, write detailed Google Flow prompts, and refine instructions for individual scenes.
Created specific prompts for camera movement, shot composition, lighting, human actions, object physics, and motion control.
Developed continuity instructions to help maintain consistent branding, colours, products, equipment, and environments across clips.
Refined prompts in response to generation issues and client feedback.
3. AI Image Generation & Visual References
Used ChatGPT for image generation and visual reference development.
Created vertical reference images for individual scenes before video generation.
Used reference images to guide product placement, coffee preparation, customer interactions, and brand-focused compositions.
Worked to maintain Café Joyeux's yellow-and-black palette, original mascot style, packaging appearance, and visual consistency.
4. AI Video Generation
Used Google Flow to convert reference images and detailed prompts into individual video clips.
Generated 8-second source clips covering coffee preparation, espresso extraction, latte art, customer experiences, café interactions, product displays, and logo animation.
Iteratively refined prompts to improve movement, realistic physics, transitions, and consistency between scenes.
Focused on natural hand movements, realistic coffee preparation, smooth camera motion, and controlled product animation.
5. Professional Video Editing in Adobe Premiere Pro
Imported and organised the generated video clips in Adobe Premiere Pro.
Reviewed the footage and selected the relevant portions for the final sequence.
Trimmed and arranged clips in the intended narrative order.
Joined the separate AI-generated clips into one cohesive promotional video.
Refined pacing, scene transitions, visual continuity, and overall presentation.
Prepared the assembled video as a unified brand-focused advertisement.
Production Workflow
Step 1 — Research & Planning: Reviewed the brand's public website, visual identity, coffee products, and available reference material.
Step 2 — Storyboarding: Planned the scenes and defined the visual purpose of each clip.
Step 3 — Image Creation: Generated scene-specific reference images and established the intended composition.
Step 4 — Prompt Engineering: Wrote detailed Google Flow prompts specifying actions, motion, physics, brand continuity, and camera behaviour.
Step 5 — AI Video Generation: Generated individual 8-second clips using Google Flow and supplied reference images.
Step 6 — Review & Refinement: Reviewed generated results and adjusted prompts where movement, branding, or visual consistency needed improvement.
Step 7 — Final Editing: Used Adobe Premiere Pro to trim, sequence, join, and refine the clips into one complete promotional video.
Tools & Their Purpose
ChatGPT
Creative planning and storyboarding
Prompt engineering for Google Flow
AI image generation
Visual direction and continuity instructions
Refinement of scenes and motion requirements
Google Flow
Image-to-video generation
AI-generated promotional footage
Camera and object movement based on reference images and prompts
Iterative clip generation and visual experimentation
Adobe Premiere Pro
Video editing and clip trimming
Scene sequencing and assembly
Transitions and pacing
Visual continuity and final presentation
Creative Direction
The advertisement combines realistic coffee-making scenes with Café Joyeux's distinctive brand-inspired illustration style.
Key visual elements include:
Yellow-and-black brand colours
Mascot-inspired drawings and playful coffee doodles
Coffee beans, grinders, espresso machines, and latte art
Branded cups and coffee packaging
Warm café lighting and lifestyle-focused scenes
A dedicated product showcase and logo-led ending
Particular attention was given to realistic physics and movement, including coffee grinding, espresso extraction, milk pouring, cup handling, customer interactions, and product animation.
Final Deliverable
An AI-assisted, vertical short-form promotional video assembled from individual 8-second generated clips and edited in Adobe Premiere Pro.
The project demonstrates my workflow across creative planning, AI image generation, prompt engineering, reference-based video generation, motion refinement, brand-focused visual storytelling, and professional post-production.
Project Focus: AI Video Production | Promotional Advertising | Brand Storytelling | Prompt Engineering | Google Flow | Adobe Premiere Pro
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
Related posts
✨ A new real-world project — Astrid Mija Contti | Biotherapy & Personal Growth
I'm happy to share a project that means a lot to me: a website designed and developed in Framer for biotherapist Astrid Mija Contti, who trusted me with her project. I'm truly grateful for that trust and the opportunity to bring her vision to life. 🤍
Because this project focuses on biotherapy and personal growth, I wanted the website to feel calm, welcoming, and harmonious from the very first visit.
🪷 The lotus flower became a central part of the visual concept, symbolizing growth, rebirth, and transformation. I originally drew the lotus myself, then used ChatGPT to transform it into a 3D visual. The subtle moving effects were created with PicLumen to bring the experience to life.
Beyond the visual design, the website also features a booking system, which I developed with the assistance of ChatGPT and Codex.
This project allowed me to combine visual storytelling, thoughtful design, and AI-assisted development into a real-world experience tailored to a client's needs.
Thank you, Astrid, for your trust! 🤍
🔗 Explore the website: https://bioterapija-z-astrid-mija-contti.framer.website/
#WebDesign #UIDesign #Framer #WebDevelopment #AIAssistedDesign #RealWorldProject
Congrats on the real-world project! Calm and welcoming from the first visit is exactly what a biotherapy site needs.
🚀 Created realistic AI-generated UGC videos for PactFresh, blending product-focused storytelling with authentic visuals to create engaging marketing content.
#PactFresh #AIUGC #AIVideoGeneration #ContentCreation
The specific mention of Higgsfield is interesting because it is not the standard tool most people assume for this workflow. Curious how you balanced the product accuracy against the natural movement in those frames to keep the UGC feel authentic.
Time Machine: One Morning Across Three Eras
I built a custom Time Machine effect that transforms the same morning across 1905, 2026, and 2080
I wanted to show how our surroundings and technology may change over time while the small human rituals we carry with us remain familiar
Using invideo Editor’s timeline agent, I created a custom shader with an editable Year control: aged film grain and sepia for 1905, a clean natural look for 2026, and cyan holographic scanlines for 2080
The agent also built the bar transitions between eras directly on the timeline. I refined the timing, adjusted the shader controls, and fixed the closing titles inside invideo Editor
The first video is the final result. The second shows the process, from importing the clips and directing the agent to testing the Year control
@invideo.io @invideo.creators
#invideocontrachallenge #invideoeditor