Howdy Logo
Image of Gustavo M.

Gustavo M.
Machine Learning/AI Engineer

Spark
Sql
Al
Java
Python
Bio

Gustavo, originally from Brasília, Brazil, pursued an academic and professional path in the software industry following his undergraduate studies. Between 2007 and 2009, he obtained a Master's degree in Computer Science while continuing his professional endeavors. He later achieved a PhD in Automation and Systems Engineering at the Federal University of Santa Catarina, which included an internship at a university in Magdeburg, Germany from 2011 to 2012. Currently, he serves as a senior machine learning researcher and data engineer, with extensive experience in artificial intelligence projects.

  • Senior Machine Learning Engineer/ Data Engineer
    10/1/2022 - Present

    Developed a framework to process ML model predictions and Shapley values in a Hadoop/Spark environment utilizing Hive, Teradata, and HDFS for storage. Gained proficiency in Pyspark and interfaced with H2O Driveless API. Managed version control and continuous integration using Bitbucket, Bamboo, and Jira. Additionally, developed a Responsible AI framework within the Hadoop/Spark ecosystem, leveraging Pyspark for implementation.

    Led a corporate learning project integrating OpenAI with Pinecone and LangChain. Created APIs using Azure Functions, ensuring seamless interactions between components. Demonstrated expertise in cloud-based solutions and functional API development within the Azure environment.

  • Machine Learning Engineer
    2/1/2015 - 3/1/2022

    Led several impactful IA projects, notably enhancing technical expertise with various machine learning frameworks and deployment tools.

    For the RHI-Magnesita project, focused on forecasting demand for refractory bricks from December 2016 to April 2017, successfully built several time series models including ARIMA and XGBoost, achieving an overall accuracy of 90%. This reduced idle stock and improved cash flow.

    In the Aqtech project from January 2018 to April 2020, spearheaded a team to detect anomalies in wind turbine engines. Defined and implemented a pipeline using MlFlow and AirFlow, and developed algorithms for sound signal fault detection.

    Contributed to Printiforia from December 2019 to March 2020 by enhancing their computer vision system for t-shirt printing defect detection. Leveraged TensorFlow and Detectron2 for object detection, improving detection accuracy from 92% to 97.2%.

    During the Mypharma2Go project between June 2021 and January 2022, developed a hybrid recommendation system based on genetic exams using TensorFlow Recommender. Managed the deployment pipeline utilizing AWS Sagemaker while fulfilling roles as both a Machine Learning Developer and Software Manager.

    For the BRK-Ambiental project from February 2022 to May 2022, designed a leak detection system for water distribution using CAE - Convolutional Auto Encoder to handle anomalies, deployed with MlFlow.

    Led the Renault project from June 2020 to October 2022, focusing on fault analysis in ECU engines. Developed deep auto encoder algorithms for detecting misfire and knock noise failures, responsible for pipeline definition and algorithm implementation.

    Demonstrated expertise in machine learning, particularly in anomaly detection, time series forecasting, computer vision, and recommender systems. Proficiency with key tools and technologies like TensorFlow, Detectron2, XGBoost, ARIMA, MlFlow, AirFlow, and AWS Sagemaker was instrumental in the successful deployment of solutions.

  • Java Software Engineer
    1/1/2008 - 12/1/2009

    Spearheaded the development of a robust middleware framework designed to seamlessly connect RFID devices with web applications. Leveraged extensive knowledge in C++ for core middleware development, ensuring efficient handling and processing of RFID data. Utilized RESTful APIs to facilitate smooth communication between the middleware and web interfaces. Integrated the use of JSON formatting to standardize data exchange, enhancing the compatibility across different systems. Applied expertise in network protocols, including TCP/IP and HTTP, to ensure reliable and secure data transmission. Employed logging and diagnostic tools such as Log4j to monitor system performance and troubleshoot issues effectively. Implemented best practices in software development and testing, utilizing JUnit for unit testing to maintain high code quality. Additionally, maintained project code using Git for version control, fostering collaborative development and efficient codebase management.

  • Java Software Engineer
    1/1/2006 - 12/1/2008

    Developed a software solution enabling Two-Phase Commit (2PC) at the application level between MS COM+ and Glassfish applications. Gained deep expertise in distributed transaction management and system integration. Utilized Java and C++ for creating seamless communication channels. Employed Glassfish's robust enterprise capabilities along with MS COM+ for managing distributed components. Implemented robust error-handling and rollback mechanisms to ensure data consistency across disparate systems. Leveraged libraries and frameworks such as JDBC for database interactions and JMS for messaging services. Demonstrated proficiency in managing transaction states, coordinating commit and rollback actions, and maintaining data integrity across distributed platforms. Ensured high availability and fault tolerance of the solution through extensive testing and optimization. Conducted performance profiling and tuning to handle large transaction volumes efficiently.

  • Java Software Engineer
    1/1/2005 - 12/1/2006

    Developed expertise in BSO - Business Service Orchestration platform and FuegoBPM. Leveraged these tools to design and implement efficient business process management solutions, effectively integrating automated workflows to enhance operational efficiency. Extensive experience in creating and managing complex business processes, optimizing resource allocation, and ensuring seamless service delivery. Proficient in technical aspects of business service orchestration, contributing to the successful deployment of scalable and maintainable BPM applications.

  • Java Software Engineer
    1/1/2004 - 12/1/2004

    Developed and maintained cutting-edge civil engineering applications utilizing Java, J2EE, and Spring Framework. Gained expertise in building and optimizing features in web applications through the use of AngularJS and React. Implemented backend functionalities using Hibernate and RESTful web services. Proficient in database management, demonstrated through the use of SQL and Oracle. Ensured high-quality code through continuous integration and deployment practices leveraging Jenkins and Docker. Enhanced application performance and scalability with container orchestration using Kubernetes. Utilized various software development tools including IntelliJ IDEA, Maven, and Gradle. Employed version control and collaborative coding practices using Git. Analyzed and resolved complex system integration issues, optimizing code for efficiency and performance. Developed strong problem-solving and analytical skills through continuous debugging and troubleshooting.

  • Software Development Engineer
    1/1/2001 - 12/1/2004

    Developed a high-performance information system specifically designed for federal government agencies, demonstrating profound expertise in Java and Spring Boot for backend development. Enhanced user experience through advanced frontend technologies including Angular and React. Implemented robust and scalable database solutions using PostgreSQL and MongoDB, ensuring efficient data management and retrieval. Utilized Docker and Kubernetes for containerization and orchestration, contributing to a seamless deployment process. Applied Agile methodologies and employed Jira for task management, fostering efficient team collaboration and project tracking. Leveraged Jenkins for continuous integration and deployment, ensuring consistent application performance and reliability. Engaged in comprehensive testing using Selenium and JUnit to guarantee system functionality and stability. Contributed to code maintenance and version control through proficient use of Git.

  • Automation and Systems Engineering at Federal University of Santa Catarina
    2009 - 2013

  • Graduate in Data Processing at Catholic University of Brasília
    1998 - 2002

  • Computer Science at Federal University of Santa Catarina
    2007 - 2009

  • Computer Science at Otto von Guericke University Magdeburg
    2011 - 2012

  • AWS Machine Learning Specialty at AWS
    12/1/2023

  • Machine Learning with Python at Coursera
    4/1/2022

  • IBM AI Engineering Professional Certificate (V2) at Coursera
    4/1/2022

  • Deep Neural Networks with PyTorch at Coursera
    4/1/2022

  • Deep Learning with Tensorflow at Coursera
    4/1/2022

  • Deep Learning Essentials with Keras at Coursera
    4/1/2022

  • Computer Vision and Image Processing Essentials at Coursera
    4/1/2022

  • General and Python Data Science at TestDome
    1/1/2022

Gustavo is available for hire

Hire Gustavo M.
Check icon

All Howdy Candidates are vetted for skills and english proficiency.