✨ Full-Stack Engineer & ML Developer
- 🚀 Founder & Lead Engineer at Aween Rayeh – serving 109,000+ monthly users across Palestine
- 💻 Specializing in Next.js, TypeScript, React, Node.js, PostgreSQL, and production ML/AI systems
- 🎓 4th-year Computer Engineering student at An-Najah National University with a GPA of 3.97/4.0
- 🏆 Full-ride scholarship recipient from Palestinian Ministry of Education
- 🤖 Expert in ML pipelines, RAG systems, fine-tuning LLMs, and scalable cloud architecture (AWS Lambda)
- 🔧 Strong foundation in algorithms, system design, data structures, and production-grade engineering
- 📊 Built ML systems with 96% accuracy, platforms serving 100K+ users, and recommendation engines with 93% variance retention
- 🌟 Passionate about solving real-world problems through technical innovation and rigorous engineering
- 🔗 Check out my portfolio and projects below!
- Serving 109,000+ monthly users across Palestine with critical checkpoint status updates
- Built with Next.js, Supabase, and React Native (mobile)
- Production ML pipeline on AWS Lambda with 96% accuracy using fine-tuned 7B Mixtral model
- 63% cost reduction through intelligent heuristic optimization
- Real-time data extraction from Telegram channels with dynamic intervals
- Processing 8K+ songs with 223-dimensional audio features (MFCCs, VGGish embeddings)
- 93% variance retention via 4 clustering algorithms (K-Means, GMM, Hierarchical, DBSCAN)
- 73% improvement in user preference over baseline (A/B tested with 15 participants)
- Context-aware LLM playlist generator with 86% satisfaction rate
- Full-stack Next.js app with Turso backend and natural language search
- Achieved 99.5% train/test parity across 24.5M samples
- 15-40% throughput improvement over Minstrel-HT baseline
- 5-8× improvement in rare-class recall with sub-1ms latency
- Memory-efficient reservoir sampling processing 38M rows in constant 20MB
- Reduced class imbalance from 100:1 to 20:1
💡 Open to collaborating on impactful full-stack and ML projects