A single-page portfolio site showcasing six deployed AI/ML projects, each presented as a "model card" — a spec-sheet-style summary (task, architecture, dataset, stack) rather than a generic project tile.
Live site: portfolio-kbaub2rhq-ali-faraz-pys-projects.vercel.app
Every project card links to both a live, working demo and its public source code. The design borrows from real ML documentation conventions (model cards, spec sheets) rather than a typical "portfolio template" look — paper-white background, monospace labels, a small animated neural-network diagram in the hero, and category-coded accent colors (vision, generative, health, NLP, finance).
- Deepfake Detector — PyTorch, EfficientNet-B0, FastAPI, Next.js
- Picassify — Neural style transfer, PyTorch, VGG19, Streamlit
- NeuralLens — Image classification, MobileNetV2, FastAPI, Next.js
- Diabetes Detector — Random Forest, scikit-learn, Streamlit
- SentimentSense — Aspect-based sentiment analysis, DistilBERT + BART
- AetherQuant — Crypto trend classification, XGBoost
- Next.js / React
- Tailwind CSS (v4)
- Custom fonts: Space Grotesk, JetBrains Mono, Inter
- Deployed on Vercel
npm install
npm run devOpen http://localhost:3000 to view it.
portfolio-hub/
└── src/app/
├── page.js # All project data + page layout
├── layout.js # Font setup
└── globals.css # Color theme, animations