EXPERIENCE:

Hiberus Jul. **** - Present

• Data engineer at Santander bank.

Design and implement advanced PySpark pipelines, which were rigorously tested, significantly enhancing internal business processes through the integration of serverless EMR, managed Airflow, and Apache Iceberg, ensuring scalable and efficient data handling. Successfully executed migrations from PL/SQL(Oracle) to PySpark, thereby improving performance and operational efficiency. Bluetab Jun. 2022 – Jul. 2023

• Data engineer at BBVA bank.

Designed and implemented several Spark projects in Scala pipelines for internal business processes on the DATIO platform - I also became familiar with Artifactory, Dataproc and BitBucket for version control.

TMC Jan. 2022 – Jun 2022

• Data engineer at Airbus.

Designed and implemented PySpark pipelines and reporting dashboards to help monitoring the daily spare parts demand and supply, with the Supply Officer as main stakeholder.

• Machine Learning engineer at Inditex.

Helped implementing a multilanguage framework for Python, Scala, Java and R to enable Data Science teams to deploy ML models.

InAtlas Jan. 2020 – Jan 2022

• ETL/ ELT and Database migration.

Implemented various scripts in Python and performed the DB migration of a project’s database from Postgres to Snowflake, improving the overall performance substantially. My responsibilities included cleaning up and organizing data as well as implementing various ETL pipelines using Python, Talend and Snowflake.

• Footfall modelling.

Developed a model for calculating the footfall of every Spanish building’s front entrance segmented by time band. I have modeled and implemented the model using Voronoi diagrams, isochrones and shortest-pass algorithms, storing the results in a Snowflake datawarehouse . It was developed in Python, Snowflake and deployed on AWS.

• Modelling with geospatial variables through Voronoi mesh, isochrones and shortest-path algorithm.

To carry out geospatial calculations used the geovoronoi, geopandas, and networkx libraries. In order to speed up the processing came up with and implemented a custom parallelization using Python’s multiprocessing libraries and deployed to 6 EC2 instances.

• Statistical modelling for tourist density by administrative division. Developed a statistical model to estimate the average tourist density, both daily and annually, per administrative division, calculating indicators related to tourists’ expenses, number of tourists in private accommodation, hotels, etc.

UNIVERSITY PROFESSOR

Technological University José Antonio Echeverría (ISPJAE) Sep. 2008 – Jun. 2010

• Professor of Classical Mechanics I, Electromagnetism I, Mathematical Analysis and Calculus I, II, III, IV.

SKILLS:

Python

Scala

AWS

Linux

Docker

Statistics

Spark

GIT and BitBucket

DBMS (PostgreSQL, Snowflake,

Oracle)

EDUCATION:

Data Science Bootcamp

Neoland

Bachelor in Physics

Universidad de La Habana (UH)

LANGUAGES:

Spanish (Native)

English (B2)

Javier Fernández Castellanos ad4ifz@r.postjobfree.com Data Scientist / Data Engineer www.linkedin.com/in/javier-fernandez-castellanos