Alexandre Tavares is a Machine Learning Engineer and Ph.D. candidate in Electrical Engineering specializing in anomaly detection for rotating machinery, combining six years of industrial and research experience with Petrobras and Logspace. He designs and deploys unsupervised and hybrid ML systems—ranging from autoencoders and isolation forests to specialized neural networks with dynamic loss functions—that delivered an average F1 of 0.90 in industrial defect detection. With a background in biomedical engineering and an M.Sc. in AI for system identification, he bridges signal processing, time-series forecasting, and production-ready ML using TensorFlow, PyTorch and Databricks. Alexandre actively contributes to open-source AI tooling as a back-end developer on Langflow, improving vector store and document-loader flexibility for RAG and multi-agent apps. He pairs rigorous academic work with practical product impact, notably building fraud and forecasting systems for large Brazilian enterprises and developing a Python library to simplify time-series workflows.
6 years of coding experience
4 years of employment as a software developer
Bachelor in Biomedical Engineering, Engenharia Biomédia, Bachelor in Biomedical Engineering, Engenharia Biomédia at Université Claude Bernard Lyon 1
PhD in Electrical Engineering, Anomaly detection in rotating machines using Machine Learning, PhD in Electrical Engineering, Anomaly detection in rotating machines using Machine Learning at Federal University of Uberlandia (UFU)
Bachelor in Biomedical Engineering, Bachelor in Biomedical Engineering at Federal University of Uberlandia - UFU
PhD, Machine Learning, PhD, Machine Learning at Villanova University
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
Role in this project:
Back-end Developer
Contributions:1 review, 5 PRs, 10 pushes in 1 year 3 months
Contributions summary:Alexandre primarily contributed to the back-end logic of the Langflow project, focusing on adding support for extra fields in the `VectorStoreFrontendNode` and implementing the `add_extra_fields` method for various document loaders. They also introduced the necessary methods to add fields to templates and refactored code to include document loader fields. These changes indicate an effort to enhance the flexibility and functionality of the Langflow framework, specifically related to vector stores and document loading capabilities.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.
Request Free Trial
Alexandre Tavares - Machine Learning Engineer at Logspace