Summary
Fagner Cunha is a Machine Learning Engineer and PhD candidate with a decade of experience turning computer vision research into production-ready solutions, currently applying his expertise at Nubank. He has a strong track record in fine-grained and long-tail classification problems, having built large biodiversity datasets and models for insect and butterfly identification in collaborations with Mila and eButterfly. His work uniquely blends embedded systems and model optimization—deploying efficient Vision Transformer, EfficientNet, and MobileNet variants on resource-constrained devices such as Raspberry Pi–based camera modules. Fagner’s projects tackle real-world domain-shift challenges in camera-trap data and integrate biological priors like geographic distributions to improve accuracy. With roots in embedded Linux and FPGA development, he brings low-level systems intuition to ML deployment and dataset engineering. Based in São Paulo, he combines academic rigor from UFAM with practical engineering that bridges field ecology and scalable AI.
10 years of coding experience
9 years of employment as a software developer
Technician, Electronics, Technician, Electronics at Instituto Federal de Educação, Ciência e Tecnologia do Amazonas - IFAM
Doctor of Philosophy (Ph.D.), Informatics, Doctor of Philosophy (Ph.D.), Informatics at Universidade Federal do Amazonas
English, Portuguese