Philippe Remy
Co-Founder - AI ML at Skysense
Tokyo, Japan
Summary
🤩
Rockstar🎓
Top SchoolPhilippe Remy is a co-founder and AI/ML engineer based in Tokyo with 11 years of experience building deep learning systems and startups. Trained at Imperial College London with a MSc in Mathematical Statistics and Probability (Highest Distinction), he blends rigorous mathematical foundations with hands-on engineering across speech, attention, TCNs and time-series models. At Skysense he focuses on end-to-end deep learning solutions, and his open-source contributions include well-used projects like deep-speaker, keras-tcn and N-BEATS implementations, where he often improves core data pipelines, layer implementations and model training regimes. Comfortable across backend engineering, feature extraction and model internals, he pairs a globe-trotting background (Paris–London–Tokyo–LA–Bangkok) with practical product building and a knack for making complex ML components robust and production-ready.
11 years of coding experience
Master of Science (MSc), Mathematical Statistics and Probability, Highest Distinction, Master of Science (MSc), Mathematical Statistics and Probability, Highest Distinction at Imperial College London
Japanese, Spanish, French, English
Stackoverflow
1,821reputation
466kreached
6answers
4questions
keras
top-5%
machine-learning
top-5%
tensorflow
top-1%
python
top-5%
Github Skills (35)
python10
attention-mechanism10
machine-learning10
keras10
lstm10
time-series-forecasting10
deep-learning10
tensorflow10
mfcc10
neural-network10
nlp10
mfc10
data-structure9
named-entity-recognition9
pytorch9
Programming languages (14)
Github contributions (5)
philipperemy/stanford-openie-python
Jul 2016 - Jul 2022
Stanford Open Information Extraction made simple!
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:83 commits, 52 PRs, 93 pushes in 6 years 1 month
Contributions summary:Philippe primarily contributed to the development of a Python wrapper for Stanford Open Information Extraction (OpenIE). Their work included implementing core functionalities for processing text, extracting relations, and generating graph visualizations. Furthermore, the user integrated the OpenIE functionality with a library, and updated dependencies for improved functionality. This involved significant interaction with the Stanford CoreNLP library, demonstrating an understanding of NLP techniques and related tooling.
nlpextractionstanfordopen-information-extractionpython-wrapper
philipperemy/n-beats
Nov 2019 - Jan 2023
Keras/Pytorch implementation of N-BEATS: Neural basis expansion analysis for interpretable time series forecasting.
Role in this project:Backend Developer
Backend DeveloperContributions:164 commits, 44 PRs, 160 pushes in 3 years 2 months
Contributions summary:Philippe's contributions focused on implementing and modifying the core model for the N-BEATS time series forecasting project. Their work included adding the `NBeatsNet` class, implementing block creation, and defining the `linear_space` function. They also modified the `compile_model` to configure the optimizer. The code changes clearly demonstrate their understanding of the Keras framework for building and training time series models.
pytorchforecastingdeep-learningseries-forecastingneural-networks
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.