Experience

Scanbuy

• United States
• Advertising Services
• 1 - 100 Employee

• Senior Data Engineer | Senior Machine Learning Engineer

  • Jan 2021 - Present

  INDUSTRY- Creation of Identity Graphs (DMP/DSP/CDP) with a focus on CPG and retail data.ACHIEVEMENTS- Design and implementation of a serverless machine learning pipeline, leveraging AWS and state-of-the-art algorithms (UMAP, HDBSCAN, etc.) to identify behavioral cohorts in massive datasets, enabling explainable modeling.Data governance is a key driver of this pipeline, allowing us to comply with DPAs.- Greatly increase the quality of a behavioral data source through statistical analysis.- Successfully lead technical integrations with business partners.- Guide my team to follow best practices to reduce technical debt.- Mentor co-workers so they can reach technical maturity (kudos to them!).SKILLS- Python, Spark, AWS (S3, EC2, EMR, Lambda, CloudWatch, Step Functions, SageMaker), Serverless, LocalStack, Machine Learning, SQL, Testing, Airflow, Jenkins, Parquet, Bash, Git,



M&GT Consulting

• Spain
• Information Technology & Services
• 1 - 100 Employee

• Big Data Engineer | Technical Lead

  • Jul 2019 - Jan 2021

  CLIENT- Major international bankACHIEVEMENTS- Design & implementation of the application architecture following SOLID principles in accordance with business needs. This architecture allowed fast value delivery to the client and managed to reduce the cognitive load of the developers.- Tuning of the application interaction with the cluster to enable a fair use of resources and improve performance & reliability.- Release management & deployment of the application.- Establishment of good practices and tutoring.SKILLS- Hadoop Ecosystem (HDFS, YARN, Hive) on CDH, Spark, Python, Jenkins, Control-M, SQL, Git, Bash, Jira, Confluence, Avro, SQL, Git, Bash, Jira, Confluence.

• Machine Learning Engineer | Data Engineer

  • Jan 2019 - Jun 2019

  CLIENT- Major international transportation infrastructure operatorACHIEVEMENTS- Implementation of the CRISP-DM process to design, develop and deploy into production an on-demand multi-class classifier, solving a complex use case with imbalanced/categorical data and complying with strong run-time restrictions. The new system increased the precision by ~40% over the previous one, which allowed to automate almost an entire department.- Development & revision of an ETL workflow to integrate many heterogeneous data sources into a Data Warehouse, greatly aiding decision-makers.SKILLS- Python, Machine Learning, AWS (S3, EC2), Elasticsearch, Dremio (DaaS), PostgreSQL/PostGIS, PL/pgSQL, Graphs (NetworkX, OSM, osmnx), Excel/PDF custom parsing, Tableau, REST API, Git, Redmine.



Big Data Health Technologies - BDH Techno

• Spain
• IT Services and IT Consulting
• 1 - 100 Employee

• Data Scientist

  • Dec 2017 - Mar 2018

  CLIENT- Medical researchersPROJECT- Early identification of breast cancer from metabolomic dataACHIEVEMENTS- Use of stacked ensembles and oversampling techniques to improve the accuracy and increase the sensitivity over the previous system. The objective was to forecast the health conditions of the patients (prediction) and identify the most relevant metabolites (inference).- Design of multiple experiments and reports.SKILLS- Python, Machine Learning, Jupyter Notebook, Weka, LaTeX.



Research Group RFIDUJA

• Undergraduate Research Student

  • May 2016 - Oct 2016

  - Self-taught study of statistical learning theory- Design and development of a Web Scraper- CMS maintenance - Self-taught study of statistical learning theory- Design and development of a Web Scraper- CMS maintenance