In this example, I wanted to implement concepts of Computer Vision.

I developed this web app for a client who needs to estimate how many new release books will sell in the first 12 months. I used historical sales data for books published in Argentina across several years and external data regarding the specifications of books and authors. The trained model was an LGBMRegressor classifier.

This app shows several socio-demographics indicators from Avellaneda city. It was developed using census radios information produced by The National Institute of Statistics and Censuses (INDEC) in the 2010 Census. Census radios are geographical units that have, on average, 300 homes in the cities. Currently, Argentina has 51.408 census radios. The app was built using Shiny.

With Time Series API published by Ministry of Modernization I’ve identified a few relevant economic indicators to help assess the current economic trend. The app was built using Dash and Plotly in Python, and deployed in Heroku

Analysis of private sector debtors in Argentina

I have a B.Sc. in Economics and an M.Sc. in Finance; I worked in the capital market as an equity analyst and investment manager. I also worked as a supervisor in the National Social Security Administrator (ANSES) to execute several projects. I became interested in Data Science and programming and joined ITBA for a Specialization Course in Data Science. I finished this degree in 2019. Since then, I have trained myself and worked on several personal and professional projects. In 2021, I became a Machine Learning Engineer at Fourthbrain in the US.