Agustín Azzinnari
Senior Software Engineer at Consumer Acquisition
Montevideo, Montevideo, Uruguay
Summary
🤩
Rockstar🎓
Top SchoolAgustín Azzinnari is a senior software engineer and independent ML consultant with 11 years of experience designing, building, and deploying end-to-end systems that leverage computer vision, NLP, and statistical models. He has led consulting teams at Tryolabs and now drives ML initiatives at Consumer Acquisition, combining hands-on engineering with product-minded strategy and client-facing collaboration. His background spans the full stack and infrastructure, which enables him to scope MVPs, evaluate feasibility of critical components, and shepherd models into production. An active open-source contributor, he improved evaluation tooling in the Luminoth object-detection toolkit to surface per-class AP and internal losses for deeper model insights. Based in Montevideo, he regularly works with international clients remotely and has delivered workshops and technical sales support across markets.
11 years of coding experience
7 years of employment as a software developer
Ingeniería en Computación, Computer Engineering, Thesis: 12/12, Ingeniería en Computación, Computer Engineering, Thesis: 12/12 at Universidad de la República
English, Spanish, Portuguese
Stackoverflow
13reputation
1kreached
1answer
1question
Github Skills (14)
computer-vision10
eval10
deeplearning-ai10
machine-learning10
deep-learning10
tensorflow10
python10
evaluation10
object-detection9
json6
ajax6
aquamacs6
jquery6
emacs6
Programming languages (4)
Github contributions (5)
tryolabs/luminoth
Aug 2017 - Apr 2018
Deep Learning toolkit for Computer Vision.
Role in this project:
ML Engineer
ML EngineerContributions:5 releases, 11 commits, 37 PRs in 8 months
Contributions summary:Agustín made several contributions related to model evaluation within the Luminoth deep learning toolkit. Their work included updating the Sonnet version dependency and refactoring the graph evaluation process for improved clarity and maintainability. Furthermore, the user added functionalities to include all internal losses and AP (Average Precision) per class to the evaluation summaries for more comprehensive model performance analysis. These changes focused on enhancing the evaluation process and providing more detailed insights into model behavior.
pytorchpythonvisiondeep-learningcomputer-vision
nagitsu/nabu
Mar 2015 - Jul 2016
Contributions:262 commits in 1 year 4 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.