Looking for a job

Andre Barros

@andrebarros

Data Engineer at Marlabs Brasil, processing ~100 TB daily for Serasa and cutting legacy pipeline time from 48 hours to 4.

Brazil

Message

At Marlabs Brasil, I worked on data platforms for Coca-Cola and Serasa. For Serasa, I processed approximately 100 TB of data per day and migrated legacy Mainframe pipelines to Spark on AWS EMR and Databricks, reducing processing time from 48 hours to 4.

Previously, at Toro Investimentos, I created and maintained Data Warehouses and Data Lakes for internal teams. At TGR, I built XGBoost models that generated KPIs with 80% accuracy to support proactive network and field-service interventions.

Experience

Work history, roles, and key accomplishments

Current

Data Engineer

Current

Marlabs Brasil

Mar 2023 - Present (3 years 8 months)

Client: Coca-Cola | 01/2026 - 07/2026 | Databricks, Microsoft Analytics Services, Power BI

• Built raw data ingestion pipelines based on a medallion architecture using Databricks, designed for high data volumes and clear organization across data layers.
• Optimized the analytics layer and Power BI consumption by structuring dimensions and calculations in Microsoft Analytics Services, improving

Data Engineer

Toro Investimentos

Feb 2020 - Feb 2023 (3 years)

Stack: Airflow, AWS, GCP, SQL Server, Oracle, Python, Spark

• Led the creation and maintenance of Data Warehouses and Data Lakes for internal consumption across multiple areas of the organization.
• Orchestrated data pipelines with Apache Airflow in an on-premises environment and developed ETL processes using Python and Spark.
• Integrated data from SQL/NoSQL databases, APIs, FTPs, and Mainfra

Data Engineer

TGR - Tecnologia em Gestão de Redes

Jan 2019 - Dec 2019 (11 months)

Stack: Airflow, AWS, PostgreSQL, Python, Spark, XGBoost

• Developed and managed ETL processes with Python and Spark, orchestrated by Apache Airflow.
• Built a Data Warehouse to support Data Science and BI teams, increasing data availability for analytics and reporting.
• Built XGBoost models to estimate data quality and network signal degradation, integrating telemetry through the TR-069 proto

Data Developer

Axiros

Jun 2017 - Oct 2018 (1 year 4 months)

Stack: AWS, MongoDB, Python

• Developed ETL processes in Python to collect data through APIs and the TR-069 protocol.
• Stored and processed large volumes of data in MongoDB, creating aggregations for analytics and KPI dashboards.
• Implemented data solutions that enabled detailed visualizations and actionable insights for business and technical teams.

Teacher

Colégio Chalupe

Sep 2014 - May 2017 (2 years 8 months)

Teach physics to elementary and high school students.

Scientific Researcher

IPEN

Oct 2012 - Oct 2014 (2 years)

Scientific initiation working with X-ray diffraction and material analysis.
Python Developer. Study of social network and machine learning.

Technical Support Specialist

Instituto de Química - Universidade de São Paulo

Nov 2010 - Nov 2012 (2 years)

Trainee for 2 years, working with servers and support to users.

Governança de dados: garantindo a qualidade de dados com a biblioteca Pydeequ

Dec 2025

Univesp Oficial

Bachelor's degree, Computer Science

2019 - 2023

The course prepares the scientist to work with mathematical, statistical and computational tools aimed at helping in the process of extracting knowledge from data, which helps in decision making.

Fatec Carapicuiba

Bachelor of Technology (B.Tech.), Computer Systems Analysis

2015 - 2018

Development and learning of digital games for application in education, entertainment and business development.

USP - Universidade de São Paulo

Master's degree, Nuclear Physics

2015 - 2017

Study of microstructural properties by X-ray diffraction analysis

Universidade de São Paulo

Physics

2009 - 2015

Scientific research at the Institute of Nuclear Energy (IPEN), for the period 2013 to 2014 in the area of materials research through crystallography and research on knowledge networks.