Developed benchmarks to evaluate Musical Source Separation (MSS) models using deep neural networks. These benchmarks were utilized to enhance the Moises/Music.ai MSS model.

As the Leader of Trainees in a department dedicated to teaching software development to newcomers at TAIL, I conduct classes on data science, machine learning, and general software engineering. My goal is to empower these individuals to develop exceptional projects and grow into skilled professionals.

TAIL (Technology and Artificial Intelligence League) is a non-profit organization that connects people and fosters knowledge in data science and machine learning. The league focuses on building projects that engage members in understanding how AI models can enhance performance and improve tasks.

At TRIL, I develop features and applications to help companies improve performance and make better use of raw data in third-party services. My primary focus is on developing LLMs to answer user queries and creating Machine Learning Models to assist these companies in making informed decisions.I also conduct research on the impact of Convolutional Neural Networks (CNNs) and Deep Neural Networks (DNNs) in healthcare, exploring how these models can improve diagnosis and make classifications more efficient

I developed a feature for the company called 'Drica AI', an application that gathers information from clients databases more quickly and uses AI to predict insights. This feature was designed to improve the efficiency of consulting information, allowing users to send text messages to retrieve data instead of manually performing the tasks.