Merkis Ruiz

Machine Learning Engineer with hands-on expertise in building end-to-end ML pipelines. Demonstrates strong proficiency in deploying production-grade systems on Google Cloud Platform, utilizing Docker and CI/CD automation to enhance operational efficiency.

Near Earth Asteroid Predictor 

End-to-end ML system predicting asteroid hazard classification, MOID and Absolute Magnitude, using NASA Near-Earth Asteroids and Close Approaches datasets.

• Built scikit-learn Pipeline with ColumnTransformer for feature preprocessing, including OneHotEncoder and StandardScaler.
• Compared several models using StratifiedKFold cross-validation, including RandomForest, GradientBoosting, LogisticRegression, and LinearRegression.
• Optimized hyperparameters via GridSearchCV to improve model performance.
• Deployed FastAPI microservice containerized with Docker on Google Cloud Platform.
• Implemented CI/CD pipeline using GitHub Actions for automated testing, building and deploying