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
My Data & AI Portfolio — Turning Ideas Into Real, Shipped Products 🚀
From Data to AI Products — My Work & Case Studies
What started as learning has turned into building real products.
I created my portfolio to showcase how I approach real-world problems using Data Analytics, Machine Learning, AI, and software development.
From business intelligence dashboards and customer segmentation to predictive ML models, recommendation systems, and a live AI product, every project is focused on turning technical capabilities into something useful.
Featured work:
📊 BI & Analytics
🤖 Machine Learning
🌐 AI Web Applications
📱 Mobile Apps
Featured projects:
• Diabetes Risk Detection
• Stock Market Analysis
• Netflix Content Analysis
• Customer RFM Analysis
• Music Recommendation System
• FitFuel AI Guru
Available for freelance work:
Websites • AI Apps • Dashboards • ML Solutions • Mobile Apps
Portfolio → https://datascientist2004.github.io/Portfolio/
Have a product idea or business problem? Let’s turn it into something real.
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
Figma Football
An unofficial football concept inspired by Figma and its visual language.
The ball was created using a custom 3D generator I built in ChatGPT, with animated interface icons and graphics developed with Claude.
Part design experiment, part exploration of how custom AI-powered tools can become part of the creative process.
The interface icons translate surprisingly well into a football identity.
This UI was really great nice job Piotr🔥🔥🔥
An interaction study for Mare’s collection feed, exploring how to make groups legible within a dense, mixed-format archive.
The layout preserves image proportions and integrates collection labels into the packing system. On hover, image-derived gradients establish the collection boundary, while subtle displacement creates space around the active group without changing its position or scale.
The motion studies isolate assembly, packing, colour and proximity, using real collection content to examine timing, spatial continuity and hover behaviour.
