I'm a data scientist and machine learning engineer focused on building robust pipelines, optimizing deep learning architectures, and serving intelligent systems in production. I specialize in turning raw, messy datasets into reliable, production-ready predictive models.
- 🏆 ISEF National Finalist — built a pipeline combining custom CNN architectures with an optimized Mistral LLM implementation
- 🧪 Kaggle Competitor — rigorous feature engineering and ensemble modeling for tabular prediction tasks
- ⚙️ Core Philosophy — moving beyond notebook experiments to build scalable, production-grade AI applications
Machine Learning & Deep Learning
Includes custom PyTorch training loops, custom metrics, and end-to-end Pandas/NumPy/SciPy data workflows.
Visualization
Deployment & Model Serving
-4285F4?style=flat-square&logo=googlechrome&logoColor=white){kind=icon}
React and Chrome Extensions (Manifest V3) are used primarily for building AI-integrated interfaces and developer productivity tools.
- Advanced Architecture Tuning — fine-tuning and blending large language models (LLMs) with spatial/vision CNNs
- End-to-End Tabular Pipelines — robust preprocessing, leakage prevention, and high-performance gradient boosting ensembles for classification and regression
- Productivity Engineering — custom browser tools that reduce digital friction and optimize daily development output
- 📊 Kaggle: [profile]
- 💼 LinkedIn: [profile]
- 📧 Email: ywael000@gmail.com
- 💬 Ask me about: custom training loops in PyTorch, feature engineering strategies, or scaling ML models into production