Diego Pereira Email & Phone Number

São Paulo, Brazil

• Built a baseline algorithm for social feed sorting with the help of PySpark.• Conducted analysis and desiged rules for spam and bot activity detection making use of Python data libraries, such as pandas, numpy, matplotlib and scikit-learn.• Designed and tested different supervised Machine Learning models for spam and bad behavior detection on app’s social feed.• Partnered with product team to create a process for model monitoring and retraining.

Rio De Janeiro E Região, Brasil

• Creation, Maintenance and Continuous Improvement of Fraud Detection / Prevention System (using SQL-based batch system).• Conducted discovery using data for deep dive analysis to gather insights for fraud and bad behavior detection (using Python data libraries, such as pandas, numpy, matplotlib and scikit-learn). • Partnered with product team to create the first real-time supervised model for fraud detection on chat application.• Partnered with operation’s team to create quality processes and models monitoring.• Assisted the definition and development of products team's KPIs in Trust and Safety area.• Built operational reports in Tableau or Redash for models and rules quality tracking.

• Maintained malware and phishing database. • Conducted file analysis and its behaviours, creating rules to automate file classification.• Developed Python scripts for file processing and classification.• Wrote Python web scripts for report page.• Worked with SQL scripts and procedures to provide statistics to the report page.

• Modelled data dictionary for BI applications used by the largest entertainment company in Brazil. • Contributed in creating a method to analyse 30k lines of PL/SQL in 2 months, which improved the Oracle 10i database querying performance by 41%.• Designed the technical content for a brochure used by HP’s marketing department.• Participated in internal experiments, such as unstructured data processing and sentiment analysis on social networks using Vertica Big Data Solutions.• Supported for 3 months an artistic resources management system, which is used by the largest entertainment company in Brazil, using PL/SQL knowledge.

Information Systems, Information Technology

http://www.ic.uff.br/index.php/en-GB/undergraduate-programs/information-systems

Master Of Science Degree Course In Computer Science - Emphasis On Data Science

Business Computing - 3Rd Year, Information Technology

Science Without Borders exchange program http://www.dit.ie/catalogue/Programmes/Details/DT354?tab=Programme%20Structure