Jorge Espinoza Abud

I currently lead a small team of data scientists, providing technical guide and contributing in projects related to risk.

As a lead data scientist I help other data scientists to work out the technical details of their projects and at the same time to build their skills and become more independent.
I lead the data science projects related to collections. It's part of my job to collaborate with other stakeholders, prioritize projects, build solutions and communicate updates and results.

As an individual contributor I build models and analize data to meet specific businness goals, such as:

• Define and detect early warnings to help the collections team take action at the level of industry and specific clients.
• Income estimation model, to assign credit lines to clients with missing income information.
• Customer lifetime value, to predict the revenue from our card clients.
• Customer churn model, to take action on clients with a high probability to cancel their accounts.
• Model to estimate customer propensity to accept credit line offers, to help the sales team to prioritize.

Among many others.

I also contribute to develop internal tools to make the data science process more efficient. Specifically on steps like monitoring and debug models.

I worked on projects for 3 different squads: growth, risk and collections. Some examples:

• Time series model to analyze credit search behavior at industry level.
• Risk estimation for recurrent clients of working capital.
• Model to estimate the best date to collect money from our clients.

@san pancho (now jellyfish) I helped them with descriptive data analysis to find segments and patterns from customers data. Developed a model to forecast trends in customer behavior and optimize targeting strategies.
@teranalytics I worked mainly on a business model for a new product using monte carlo simulations, I also built a simple web platform to visualize results. The project was a proposal to help the client make business decisions according to future projections.
@blackinntech Developed an ETL pipeline and worked with unstructured data using NLP models to detect and extract useful information from text. This work was implemented to provide background checks and driver hiring insights for a big transportation service app.
@sanabria Helped on web development and IT tasks.

I worked on software development for product life-cycle management consulting team.
Main stack: Java, Oracle platform.

I was in charge of the development and maintenance of an ETL system focused mainly on real estate data, providing the raw material to the company’s pipeline and services.

Side project as a Data Scientist I worked with natural language processing models to ex- tract relevant information from real state data and build better features for predictive models in our pipeline.

Web scraping and scrapers maintenance.

I was in the design step of the production process of marine propellers. Also worked on the mechanical design and stress-strain analysis of a new product.

Education

Tec de Monterrey

BSc Mechatronics Engineering

Stack & libraries

Pandas, Numpy, Scikit-learn, NLTK, SpaCy, Tensorflow, Keras, etc.

Languages

Spanish, English, Python