Gustavo Souto

• Build machine learning models to detect anomalies/events in time‑series data using libraries like Scikit‑Learn, Pandas and Pytorch. Thus, ensuring proactive maintenance of heavy machinery, reducing downtime and operational cost savings. • Develop end‑to‑end machine learning systems using tools like MLflow, Docker and AWS (SageMaker, S3 and Lambda). • Automate data science workflows (CI/CD) with GitHub Actions and Jenkins. • Developed and maintained ETL processes from various sources.

• Developed energy disaggregation models for residential load forecasting, deployed using Databricks for large‑scale production, improving efficiency and reducing customers costs. • Predicted energy consumption for appliances using machine learning, addressing challenges of limited data granularity. • Deployed forecasting models in production environments leveraging the Databricks platform and Azure ML, ensuring high‑ quality data availability for decision‑making.

• Increased insurance customer retention by 32% through churn prediction models and actionable marketing strategies, demonstrating strong analytical skills. • Built a recommendation engine for insurance products, driving a 14% increase in sales, showcasing proficiency in data modeling principles. • Streamlined data ingestion and ETL processes, enabling efficient data exploration for the Data Science team. • Partnered closely with product directors to turn business problems into data‑driven solutions.

• Designed hands-on robotics programs for kids and teens that encouraged programming skills, teamwork, and interdisciplinary learning.

• Developed an SVM-based model to detect traffic anomalies in Dublin using SCATS sensor data, applying time-series analysis. I also integrated data from Twitter and bus sensors, which helped not just in spotting anomalies but also in understanding their root causes.

Main responsibilities: • Mentor undergraduate students (Research projects). • Lecture the following courses: • Software Engineering • Development environments and databases • Web Development • Introduction to Logic • Introduction to Informatics

(PhD not finished) • Researched machine learning models to detect traffic anomalies and uncover underlying patterns. • Carried out this work while hosted by Fraunhofer ISST as a visiting researcher.

• Built a computer vision-based system to monitor vital signs in ALS patients, helping track their health in a non-invasive way. • Received the CAPES‑REUNI Scholarship in recognition of academic excellence and research potential.

Created an educational app tailored for the Nipo-Brazilian Digital TV System, making learning more accessible through interactive content.