$ python portfolio.py
I build practical machine learning solutions using Python — from data exploration and feature engineering to model training, evaluation, and deployment-ready pipelines.
Education, certifications, and continuous learning roadmap
Universidad de los Hemisferios & EELA
TripleTen · Feb 2025 — Mar 2026
GEM EDUCA
ACTUMLOGOS
Stanford University
SoftServe
Lviv Polytechnic National University
Professional trajectory across IT, AI, and customer success
Background & context that shaped my ML journey
With 8 years of experience in IT, including roles in technical support and technical customer success, I've developed a deep understanding of how technology serves real users and real business problems.
That hands-on experience taught me to think systematically, communicate clearly with technical and non-technical stakeholders, and always focus on outcomes over tools.
Over time, I became increasingly drawn to solving problems using data. What started as curiosity turned into a deliberate career transition into Data Science and Machine Learning.
I've invested the past year in rigorous study and applied projects — building classification models, regression systems, NLP pipelines, time series forecasting, and deep learning solutions.
My focus is on building practical, production-minded ML systems — not just training models, but understanding the full lifecycle: data cleaning, feature engineering, model selection, evaluation with proper metrics, and deployment considerations.
I bring a rare blend of IT experience and fresh ML skills, bridging the gap between engineering and data science.
How I approach Machine Learning problems
Define objectives, success metrics, and constraints before touching data.
EDA to understand distributions, correlations, and anomalies.
Handle missing values, outliers, and data quality issues systematically.
Create meaningful features that capture domain knowledge and patterns.
Compare algorithms based on problem type, data size, and interpretability needs.
Use proper metrics (F1, RMSE, AUC) with cross-validation to ensure reliability.
Hyperparameter tuning, feature selection, and ensemble strategies.
Model serialization, inference speed, monitoring, and maintenance planning.
Trained models and applied ML experiments
Built a classification model predicting customer churn for Beta Bank using 10,000 clients. Addressed class imbalance using SMOTE and class weighting. Compared Logistic Regression and Random Forest.
Predicted oil reserves using Linear Regression and used bootstrap simulation to estimate profit distribution across regions.
Predicted gold recovery rates in mining purification stages using multiple process parameters.
Implemented ML tasks including similarity search (kNN), custom Linear Regression using matrix algebra, and data obfuscation using invertible matrix multiplication.
Compared six regression models to predict used car prices from 350k listings.
Predicted hourly airport taxi demand using time series modeling and feature engineering.
Built an NLP pipeline using TF-IDF and Logistic Regression to classify IMDB reviews.
Built a CNN using ResNet50 transfer learning to predict age from images.
One year of structured ML experimentation
Core Python, Pandas, NumPy, Matplotlib. Data wrangling and exploratory analysis fundamentals.
Statistical foundations, hypothesis testing, supervised and unsupervised learning algorithms.
End-to-end projects: churn prediction, price estimation, risk analysis with real-world datasets.
Neural networks, CNNs, transfer learning with TensorFlow and Keras. Image classification and regression.
Text preprocessing, TF-IDF vectorization, sentiment analysis, and NLP pipeline construction.
Model optimization, deployment strategies, API serving, monitoring, and ML system design.
Tools and technologies in my ML toolkit
Experiment results from the lab notebook
Let's discuss data, models, and opportunities
$ python contact.py --name="Javier"
>>> Connecting to Javier's ML Lab...
>>> Status: Open to opportunities
>>> Interest: Python Developer | Data Science | Machine Learning roles
$ _