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
I built an AI system that detects 38 plant diseases with 99.98% accuracy across 87,000+ images.
It uses attention mechanisms to focus purely on disease markers, ignoring soil and background noise completely. Gemini AI acts as a pre-validation gate rejecting non-leaf images before they reach the model.
Looking for computer vision or agriculture AI projects — message me.
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
Bloom Lab is a cozy virtual plant-breeding mini game where users can choose a seed, plant it, care for it, help it bloom, and create new hybrid plants by combining traits.
The idea started with a simple thought:
What if a plant-care game could feel calm, expressive, and a little magical?
In the game, each plant reacts through mood changes, care meters, facial expressions, and small interactions. Users can water the plant, give it sunlight, watch it grow, and eventually discover hybrid plants like Glowleaf, Honeybloom, and Dreamberry.
This project was also special for me because it was my first time using Figma Make, along with other AI-assisted tools/models during the process. It was exciting, but also challenging. I had to learn how to work with prompts, fix broken flows, manage component states, refine animations, and make sure the final experience still felt intentional as a design.
Some things I focused on:
Building a soft and cozy visual style
Creating reusable components and plant assets
Designing smooth micro-interactions
Making the care screen feel alive through plant expressions and idle animations
Creating branching outcomes based on seed and hybrid choices
This project taught me a lot about designing with AI tools not just generating screens, but guiding, correcting, and shaping the experience until it felt like my own.
It was a challenging but really fun learning experience, and I’m happy I pushed myself to participate.
Check it out here: https://pad-flyer-18166722.figma.site
Excellent presentation and storytelling throughout the experience. Nicely done!
PRODIGI Studios' Internal Tool → Public Showcase
We love Rive for interactive animations. But recording them for portfolios and clients? Painful.
So we built our own internal Rive Player.
Not a hack. Not a screen recorder. A purpose-built web tool our designers actually use every day.
We're shipping many new features as well soon.
But first – we want your feedback.
If you work with Rive, what's one feature you wish existed?
And yes – the talking avatar in our video demo?
That came from our upcoming SaaS platform (launching soon).
It generates up to 10 minutes of continuous AI talking head content – lip-sync, natural pacing, no green screen.
Explainer videos. Course creators. Agencies who don't want to be on camera.
We'll share more very soon.
Youtube Video: https://youtu.be/FfcEoQsWMoU
Repost if you've ever struggled to record interactive animations.
Food Styling
AI Photography
Visual Direction
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
MagicPath
The canvas is infinite, and exploration is becoming the workflow. How are you using MagicPath?
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?