Data-Driven Solutions for SMBs | CEO at NTK Solutions | Data Science
I help small and medium-sized businesses transform their data into operational efficiency and measurable growth by building intelligent systems that solve real problems.
Data science engineer student focused on machine learning, business intelligence, and automation. Interested in model interpretability, LLM optimization, and building production ML systems.
Currently: LEAD University (Data Science Engineering)
Based: San José, Costa Rica
Open to collaborations and projects combining data science, ML, and real business impact.
Machine Learning — PyTorch, TensorFlow, scikit-learn, model deployment, deep learning, NLP, graph neural networks
Data — Python, Pandas, SQL, MySQL, PostgreSQL, ETL pipelines
Backend — FastAPI, Docker, Linux, Git
Analytics — Power BI, Tableau, data visualization
Design — Figma, Blender, Adobe Illustrator
Claude X-Ray: The Hidden Brain of Claude Code
Repo — A reverse-engineering analysis of the internal architecture of Claude Code
Stack: Python, Jupyter, TeX
Multi-stage ML architecture combining Mamba SSM, GATv2, and salience scoring for transparent classification systems.
Stack: PyTorch · Mamba SSM · Graph Neural Networks
Repo — This project implements a banking fraud detection system using Machine Learning techniques. Stack: Python · scikit-learn · Pandas · Matplotlib · seaborn · numpy
Repo - Containerized ML serving with FastAPI and Docker for scalable model deployment.
Stack: FastAPI · scikit-learn · Docker · Python
Repo — Geographic and transmission analysis of pandemic behavior in Costa Rica using government datasets.
Stack: Python · Pandas · GeoPandas · Matplotlib
Repo — EDA of Netflix's 90s catalog: genre trends, duration distributions, IMDB correlations.
Stack: Python · Pandas · Matplotlib