Data engineer intern at the Intelligent Management and Digital Synergies department at CEMEX, helping the sales team automate their workflow.

Project impact: Aid bioscientists predict an over- or under- production of proteins, a yet unavailable tool.Reduce costs by eliminating the need of synthesizing unnecessary sequences and help guide future sRNA design.How: A small dataset was compiled from varied literature. New features were generated using domain knowledge with the help of biotech engineers. Data was preprocessed and several models (Neural Network via TensorFlow, SVM, kNN and XGBoost via scikit-learn) were carefully fit, validated and ensembled to predict overfit.

Startup that aims to reduce post-treatment complications by collecting and synthesizing patient data, as well as the evolution of their condition through standarized questionnaires that are to be filled inside an app.