I work on the Data Support team, where I am responsible for automating customer support using Artificial Intelligence, Natural Language Processing (NLP) and Generative AI (Gen AI) models. My activities range from exploratory data analysis to the deployment, monitoring and continuous improvement of machine learning models.

Project Title: Comparative Analysis of Deep Learning Models for Body Fat Estimation from Photographs.In this project, I will undertake a comprehensive and meticulous analysis of various Deep Learning models with the aim of estimating body fat from photographs. The central idea will be to understand how different architectures and deep learning techniques can be efficient and accurate in the task of assessing body composition through images. Throughout the research, I will evaluate and compare the performance of different models, considering aspects such as accuracy, processing time, and robustness under different image conditions. This investigation will identify the most promising models for this specific application and highlight the main advantages and challenges associated with each of them. This work will contribute to understanding the capabilities and limitations of Deep Learning models in the health and well-being sector, paving the way for more advanced and personalized solutions in estimating body fat using imaging technology.Key Achievements: Authorship and co-authorship in scientific papers in international journals.

Project Title: Development of a System for Mobile Robot Localization in Topological Maps using Computer VisionI dedicated myself to the challenge of creating an innovative system for precise localization of mobile robots in topologically mapped environments. The developed solution focuses on the use of computer vision and artificial intelligence algorithms to enable the robot to recognize and interpret its immediate surroundings.Using PyTorch, I developed a Convolutional Neural Network (CNN) trained to identify different rooms, allowing the robot to determine its exact location within a structure. Once the room was recognized, an intelligent agent processed the optimal path, navigating autonomously through graphs that represent the environment.This project solidified my understanding of real-time applied computer vision and provided me with hands-on experience in creating intelligent agents for autonomous navigation.

Project Title: Development of a Pulmonary CT Image Segmentation System through an Approach with Morphological Geodesic Active Contour-based Level Set combined with the FoL Clustering AlgorithmDuring my tenure as a PIBITI scholar, I developed a medical image segmentation system focused on identifying and delineating lungs. Using the powerful Detectron2 tool, I was able to enhance accuracy and efficiency in analyzing these images. The system was designed to process large volumes of image data, delivering precise and quick results. To achieve this, I integrated tools like PyTorch for neural network training and OpenCV for pre and post-image processing. Furthermore, with the aid of platforms like TensorBoard, I was able to monitor and optimize model performance in real-time. This project not only allowed me to dive into the practical applications of artificial intelligence in the medical field but also strengthened my expertise in computer vision tools and deep learning.Key Achievements: Authored an article about the system in the journal: DOI: 10.21528/lnlm-vol19-no1-art4

Responsible for developing intelligent systems that elevate our clients' businesses through advanced technology. My expertise is centered on extracting value from data, creating efficient solutions, and managing machine learning models from start to finish. Main responsibilities:• Exploratory Data Analysis: Investigating and deeply understanding datasets to identify patterns, anomalies, and opportunities.• AI Model Development: Creating and optimizing artificial intelligence models, aiming for accuracy and efficiency.• Model Validation: Ensuring the robustness and reliability of models through rigorous testing.• Business Metric Analysis: Continuous evaluation to ensure the alignment of models with business objectives.• Version Management: Implementing version control practices for models and environments, ensuring fluid updates and changes.• Results Comparison: Detailed analysis of achieved results versus set goals.• Performance Optimization: Monitoring and continuous enhancement of system efficacy.• Solution Deployment: Agile implementation of systems in production environments.• Observability: Applying tools and practices to ensure real-time system visibility and monitoring.• Cloud Distribution: Ensuring scalability and availability using best cloud computing practices and tools.