Alexandre Lemos

Data Scientist with over 5 years of experience in risk analysis, data mining, and machine learning model development in the financial, industrial and logistics sectors. Expertise in deploying real-time systems and creating data pipelines, resulting in enhanced operational efficiency and optimized decision-making processes. Proficient in data visualization and KPI development, improving business insights and performance.

Senior Data Scientist at Stord (2025-09 – 2026-03)

• Built and deployed an end-to-end machine learning model (XGBoost) for Estimated Delivery Date (EDD) in Vertex AI (GCP), improving delivery accuracy by 30%+ and enhancing customer experience across the logistics platform.
• Developed a Carrier Prediction model (XGBoost) to generate early shipment signals, addressing a key upstream data gap and improving downstream EDD model accuracy by 20%, enabling more reliable and proactive delivery estimates.
• Led end-to-end technical interviews for Data Scientist candidates, defining evaluation criteria and contributing to the growth of a high-performing data team.
• Served as the ML lead within a cross-functional product team, guiding model strategy, influencing architectural decisions, and aligning stakeholders on ML-driven solutions.

Data Scientist at Trustly (2022-01 – 2025-07)

• Implemented fraud detection models using XGBoost to minimize financial losses in real-time payments, saving over $500k/year in losses for the company.
• Developed risk rules using Python and data analysis to stop fraudulent transactions in real time, saving over $300k/year in losses.
• Performed portfolio management using data analysis and data visualization to identify and manage risks across the portfolio for Trustly. Created reports and monitoring tools to present the portfolio results to stakeholders regularly.
• Built machine learning models using XGBoost and Python to be used by the Risk Operations team to assist them in prioritizing their analysis and focus on detecting fraud patterns early. Overall, fraud detection increased by over 50%.
• Developed a loss forecast model using monotonic regression, leveraging on cohort and previous months to forecast losses for the financial department.
• Created a customer segmentation model using K-Means and customer bank activity information and psychographics to understand the different needs and preferences of various customer groups and provide insights for Trustly and its customers.
• Applied AB test experiments on different occasions to see the impacts of different interventions on subsets of the population, using statistical tests to calculate the significance of the experiments.

Data Scientist at ArcelorMittal (2021-07 – 2021-12)

• Implemented machine learning models using Python and Scikit-learn, experimenting on different algorithms such as SVM, Random Forest, and MLP to detect faults in steel coil. Used SHAP analysis to understand the root causes of the fault.
• Developed an optimization algorithm using Python and statistical analysis to obtain the optimal values for a set of parameters in one industry process for Arcelor Mittal, improving performance by over 15%.
• Built machine learning models using Python, leveraging data analysis and the Random Forest algorithm to predict the temperature of steel in different stages of a process in the industry.

Data Engineer at Qi Ao Cubo (2020-01 – 2021-06)

• Developed Data Warehouse projects using dimensional modeling to feed dashboards and reports for several clients.
• Developed ETL pipelines using Pentaho to extract data from multiple sources, transform it, and load it into Data Warehouses to be used to create dashboards and reports in Tableau and Power BI for different clients.
• Created dashboards and reports using Tableau and Power BI to keep track of KPIs and produce self-serving tools for stakeholders.

Msc in Informatics – Federal University Of Espírito Santo (2024)

Bsc in Electrical Engineering – Federal University Of Espírito Santo (2020)