Mattia Di Gangi
Lead - Dubbing at AppTek.ai
Aachen, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolMattia Di Gangi is a speech and machine translation researcher-turned-engineering lead with 12 years' experience, a PhD focused on direct speech translation, and a knack for turning research into production-grade dubbing systems. Currently leading automatic dubbing at AppTek.ai, he blends hands-on software development with cross-functional coordination across engineering, ops, sales and marketing. His background includes research roles at DeepL and FBK and an AWS applied-science internship, reflecting deep expertise in speech translation pipelines. An active contributor to OpenNMT-py, he has improved training, model architectures (adding GRU support and context gates) and experiment reproducibility—showing he values both model quality and practical engineering. Based in Aachen, he prefers working on software he finds fun, bringing curiosity and product-minded rigor to complex multilingual speech problems.
12 years of coding experience
7 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at Università degli Studi di PalermoDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Università di TrentoDiploma di perito industriale capotecnico specializzazione informatica Informatica, Diploma di perito industriale capotecnico specializzazione informatica Informatica at ITIS Vittorio Emanuele III PalermoMaster's degree Computer Science , Master's degree Computer Science at Université Paris-Est Marne-la-Vallée
English, Italian, French, Russian
Stackoverflow
33reputation
830reached
0answers
2questions
Github Skills (12)
pytorch10
machine-translation10
rnn-model10
gru10
language-model10
deep-learning10
neural-machine-translation10
n10
python10
tensorflow-lite6
machine-learning6
tensorflow6
Programming languages (6)
Github contributions (5)
OpenNMT/OpenNMT-py
Mar 2017 - Jun 2017
Open Source Neural Machine Translation and (Large) Language Models in PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 8 PRs, 14 comments in 3 months
Contributions summary:Mattia primarily focused on improving the training and model architecture of the neural machine translation project. They corrected accuracy reporting formats in the `train.py` file. Additionally, the user added support for GRU cells, expanded model capabilities, and integrated context gates to enhance the neural machine translation model further. Furthermore, they introduced random seed usage for ensuring the reproducibility of experiments.
pytorchdeep-learningmachine-translationtranslationneural-machine-translation
mattiadg/FBK-Fairseq-ST
Mar 2019 - Jun 2022
An adaptation of Fairseq to (End-to-end) speech translation.
Contributions:21 commits, 1 PR, 15 pushes in 3 years 3 months
end-to-endspeech-translationmachine-translationtranslationspeech-to-text
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.