Facundo Lezama
Senior Machine Learning Engineer at Tryolabs
Montevideo, Montevideo, Uruguay
Summary
🤩
Rockstar🎓
Top SchoolFacundo Lezama is a Senior Machine Learning Engineer based in Montevideo with an MSc in Electrical Engineering and four years of hands-on experience building computer vision and video analytics systems for challenging, occlusion-heavy environments. At Tryolabs he bridges research and production, contributing to the open-source Norfair tracker while deploying and operating models on GCP (Cloud Run, Vertex AI, Pub/Sub, BigQuery) to deliver MLOps-driven impact. He has practical edge experience integrating deep segmentation and detection models on NVIDIA Jetson hardware for autonomous and smart-city applications. Originally trained in telecommunications and embedded systems, he pairs low-level engineering instincts with applied AI—optimizing real-time tracking performance and efficient production pipelines.
4 years of coding experience
7 years of employment as a software developer
Bachelor's degree, ENGINEERING, Bachelor's degree, ENGINEERING at Colegio InglésMaster of Science - MS, Electrical, Electronics and Communications Engineering, Master of Science - MS, Electrical, Electronics and Communications Engineering at Universidad de la RepúblicaMaster's degree, Social Innovation, Master's degree, Social Innovation at Learning by Helping
Spanish, English
Github Skills (4)
python10
computer-vision9
numpy8
opencv8
Programming languages (1)
Github contributions (4)
tryolabs/norfair
Feb 2022 - Jan 2023
Lightweight Python library for adding real-time multi-object tracking to any detector.
Role in this project:
ML Engineer
ML EngineerContributions:32 reviews, 40 commits, 27 PRs in 11 months
Contributions summary:Facundo significantly contributed to the `norfair` library, focusing on enhancing its multi-object tracking capabilities. Their commits introduced features for tracking objects with different classes, drawing labels, and adding embeddings to the detections. The user also incorporated camera motion tracking and updated demo files to utilize vectorized distance functions for optimized performance. These changes demonstrate a focus on improving the core functionality and efficiency of the object-tracking library.
pythonre-idreal-timeobject-detectiondetector
facundo-lezama/facundo-lezama.github.io
May 2023 - Jan 2024
Contributions:22 pushes, 4 branches in 8 months
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.