Nicolas Mengochea Email & Phone Number

Full-time intensive program for talent training in Machine Learning (acceptance rate <2%).In-depth development of several projects:• Image Classification for E-commerce: applied transfer-learning over ResNet50 CNN to create a model capable of classifying 196 different car classes, using a very small dataset for pre-processing, augmentation and training. Said data augmentation included utilizing Detectron2 for boxing the main vehicle in each image and cropping the… Show more Full-time intensive program for talent training in Machine Learning (acceptance rate <2%).In-depth development of several projects:• Image Classification for E-commerce: applied transfer-learning over ResNet50 CNN to create a model capable of classifying 196 different car classes, using a very small dataset for pre-processing, augmentation and training. Said data augmentation included utilizing Detectron2 for boxing the main vehicle in each image and cropping the background.• Sentiment analysis of movie reviews: manipulated data that is not in a traditional format, pre-processed it, vectorized text data using BoW and TF-IDF. Trained a word embedding and used it as a vectorization source for the data. Trained a sentiment analysis model to detect positive and negative opinions for movie reviews.• Home Credit Risk Analysis: predicted whether a person applying for a home credit will be able to repay its debt or not. Manipulated and visualized data, and performed data pre-processing for a large dataset of +350,000 transactions. Trained many supervised models achieving +0.75 ROC AUC. Models used were DecisionTree, XGBoost, and LightGBM.• Salary Prediction Model: The goal was to predict salary levels based on historical data for sports players. Collected and analyzed data via an API using Python and Pandas. The original data was unbalanced. Cleaned up data, generated additional fields and stored and created a base dataset. Manipulated and visualized data. Performed feature engineering and standardization. Selected evaluation metrics and baseline models. Trained a linear regression model, achieving an F1 score of 76%. Show less

Strategic planning for the whole lifecycle of a large category of products.Achieved a raise of 10% in Net profits of the last category I had in charge, achieved by proper sales forecasting, purchasing and stock control, optimum product mix, and measured discount schedules.This accurate decision-making was possible by the improvement of data gathering, sourcing missing data, and building new models to turn data into information.

Direct report to the Marketing Director. Analysis, control, and creation of KPIs and Sales Dashboards. Communication of economic and performance reports to foreign clients. Soft Skills: innovation, communication, adaptation to change. Hard Skills: Excel: macros, charts, dashboards.

Industrial Engineering, Business Administration

Machine Learning Engineer