Machine Learning Engineer

Pavise:

• Developed by Miguel Pinheiro, Pavise is a machine learning product that mutes specific words, especially profanity, in videos.
• Utilizes the ASR model Whisper for identifying and muting unwanted words, creating a family-friendly viewing experience.
• Built with cloud technologies including Cloud Run, Compute Engine, App Engine, Cloud Storage, and Cloud Tasks for scalability and performance.

Explaining ML:

• Created the "Explaining ML" series to demystify machine learning concepts.
• Covers a range of topics from Linear Regression to Neural Networks and Transformers, providing learners with a deeper understanding of machine learning techniques like GLMs, Decision Trees, and Unsupervised Learning.

• Designed and developed an automated chatbot dialog creation and editing tool, incorporating powerful abstractions that significantly enhanced productivity;
• Created an inference model to accurately predict fraudulent account sharing within Globo's streaming service, contributing to a more secure user experience;
• Spearheaded developing and maintaining an ML Platform, streamlining the training and test processes for machine learning models;
• Leveraged expertise in neural networks, PyTorch, TensorFlow, and LLM to implement cutting-edge machine learning solutions;
• Stayed abreast of the latest advancements in machine learning and technology, continuously enhancing skills to drive innovation and excellence.