Luca Di Liello
Applied Scientist at Amazon
Meran - Merano, Trentino – Alto Adige/Südtirol, Italy
Summary
🤩
Rockstar🎓
Top SchoolLuca Di Liello is an Applied Scientist with nine years of experience specializing in pre-training large Transformer models and improving performance on Question Answering and Answer Sentence Selection. He holds a PhD in Natural Language Processing from Università di Trento and blends deep academic rigor with production-focused work at Amazon Alexa AI. Luca has contributed to widely used open-source tooling — adding information retrieval metrics and robust tests to Lightning-AI/torchmetrics — demonstrating attention to evaluation fidelity across distributed PyTorch environments. His track record includes multiple applied scientist roles at Amazon and research during a PhD, giving him a strong pipeline from research experiments to scalable model development. Based in Merano, Italy, he brings a pragmatic, metrics-driven approach to optimizing NLP systems for real-world conversational AI.
9 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Natural Language Processing, Doctor of Philosophy - PhD, Natural Language Processing at Università di Trento
Stackoverflow
1,535reputation
1.2mreached
33answers
22questions
Github Skills (14)
pytorch10
machine-learning10
information-retrieval10
pytest10
python10
metric10
data-science9
deeplearning-ai9
deep-learning9
neural-network6
type-conversion6
javascript6
sorting6
react6
Programming languages (7)
Github contributions (5)
Lightning-AI/torchmetrics
Mar 2021 - Nov 2022
Machine learning metrics for distributed, scalable PyTorch applications.
Role in this project:
ML Engineer
ML EngineerContributions:48 reviews, 14 commits, 14 PRs in 1 year 8 months
Contributions summary:Luca primarily contributed to the implementation and testing of machine learning metrics within the `torchmetrics` library. They developed and integrated new metrics related to Information Retrieval, including Mean Average Precision, Mean Reciprocal Rank, Precision, Recall, Fall-out, Hit Rate, and Normalized Discounted Cumulative Gain. They refactored and optimized existing tests, improved code coverage, and ensured compatibility across different PyTorch versions, focusing on accuracy and robustness of the metrics.
pytorchscalablepythonanalysesdata-science
lucadiliello/lightcheck
Mar 2019 - Sep 2021
Contributions:2 PRs, 13 pushes, 1 branch in 2 years 6 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.