Built ETL pipelines using Python, Airflow, SQL Server and Looker Studio to automate the ETL process for the data platform gathering data to monitor the sector scenario with up-to-date data on weather, price forecasting and generation.Built and maintained an Airflow instance to automate DAGs that executed various tasks such as ETL and data replication processes, allowing the automatization of hourly tasks, to schedule and monitor data workflows efficiently.Developed a pipeline to provide business teams with information about energy measurements for clients in the energy sector and their own generator plants. The main goal was to automate the process of retrieving and processing the data from an API, and then making it available to the business teams through dashboards and Excel files, using SQL Server database, Power Query, SOAP API, XML.Determined the requirements and generated a list of potential clients for the sales team to contact, by combining data from different sources and using geospatial analysis to match the data using Python, GeoPandas, Google Cloud Platform (GCP), BigQuery, OpenStreetMap Geocoding API. This resulted in a list of over 100 potential suitable clients from different sectors, outperforming the quality of the leads provided by a third party.

Developed and evaluated strategic KPIs such as Retention, CAC, TPV, and daily active users using AWS Athena, PostgreSQL, Metabase and Looker Studio to assist the decision-making process by the product team of a client in the contactless payments industry, in areas such as marketing campaigns, client segmentation and product performanceAutomated pipelines that fed dashboards using Python, Airflow and Looker Studio to compile data from various data sources such as an S3 data lake, PostgreSQL and generate reports for the product team, decreasing the required time to create reports by 50%, going from taking two weeks to one.Analyzed credit card transactions using Python and SQL to identify trends in client behavior.All reports were compliant with brazil’s data protection law, applying data governance and data management best practices.Developed an extensive study in conjunction with the Product team on the monthly retention rates and trends for the customers of a client in the payments industry, using Excel for prototyping and Python for automation, to understand their behavior over time and how it differed amongst different client categories based on RFM (Recency, Frequency and Monetary Value); this allowed, among others, to forecast the retention for the next twelve months in advance.

Developed a crawler using Python’s Scrapy to download historical (from 2010 up to 2021) crime data from the crime statistics department in São Paulo state relative to two types of crime occurrences: homicides, car thefts/carjackings and their respective locations. Used Pandas to convert the files from Excel to parquet format in order to optimize the speed in the analysis.Leveraged Python’s GeoPandas jointly with seaborn packages to analyze spatial crime data in the city of São Paulo through the creation of maps and plots that gave a better understanding of the behavior of criminal activities in the city as well as its temporal behavior.Created dashboards to display relevant information on statistics and trends on PowerBI.

- Responsible for analysis of the PAM (Pesquisa Agropecuária Municipal) data in order to generate insights that were published in a large study.- Support on the IDEB educational data analysis project for the cities in the Sudene region.- Support on more intensive tasks such as handling and treating large datasets/files with R.- Support on data analysis using Excel and R.