I am a AI Eng & Data Guy with a proven track record in building high-performing machine learning architectures and end-to-end data systems. My solutions on Kaggle have earned global Top 1% and Top 6% finishes, community medals, and hundreds of upvotes.
I specialize in:
- Competitive Machine Learning & Ensembling: Advanced feature engineering, leak-free cross-validation pipelines, and gradient boosting optimization (XGBoost, CatBoost, LightGBM).
- Deep Learning & NLP: Bidirectional LSTMs, transformer fine-tuning, sentiment classification, and sarcasm detection.
- Generative AI & RAG: Production Retrieval-Augmented Generation architectures featuring dense vector search, cross-encoder re-ranking, and Google Gemini API integration.
- Computer Vision: Pascal VOC to YOLO annotation pipelines, fine-tuning YOLOv8 and YOLO11 for real-time object detection.
- Data Engineering & Automation: Web scraping at scale (
yt-dlp), multi-language corpus extraction, and automated ETL pipelines.
| Category | Technologies & Frameworks |
|---|---|
| Languages | Python, SQL, Bash |
| Machine Learning | Scikit-Learn, XGBoost, CatBoost, LightGBM, Yellowbrick |
| Deep Learning & Computer Vision | TensorFlow, Keras, PyTorch, Ultralytics YOLOv8 / YOLO11, OpenCV |
| NLP & Generative AI | SentenceTransformers, Cross-Encoders, Google Gemini API, NLTK, RAG Architectures |
| Data Engineering | Pandas, NumPy, yt-dlp, Beautiful Soup, OpenPyXL, Scrapy |
| Visualization & BI | Matplotlib, Seaborn, Plotly, Mermaid.js |
| DevOps & Practices | Git, GitHub Actions, Docker, Jupyter, Linux, Clean Pipeline Design |
- Kaggle: kaggle.com/lazer999
- GitHub: github.com/musaoc
- LinkedIn: linkedin.com/in/musaoc
Open to Machine Learning Engineering, Data Science, and AI consulting roles.