Sara Rabhi
Senior Applied Research Scientist at NVIDIA
Vancouver, British Columbia, Canada
Summary
🤩
Rockstar🎓
Top SchoolSara Rabhi is a Senior Applied Research Scientist at NVIDIA with a decade of experience applying NLP, sequential modeling, and ML theory to real-world problems. She holds a PhD focused on multimodal patient pathway modeling and has driven performance-focused work on recommender systems, including notable contributions to NVIDIA-Merlin’s Transformers4Rec library integrating Transformer architectures for session-based recommendation. Her background spans industry R&D and internships at Twitter and NVIDIA where she optimized large-scale ETL and accelerated recommender training (15x) using RAPIDS and PyTorch. Sara brings practical production skills—model engineering, refactoring, and test-driven features—alongside strong academic rigor from top French institutions. She also has a track record of translating models into business impact, from predicting product launch sales at Louis Vuitton to fraud detection and competitive Kaggle work. Based in Vancouver, she blends deep research instincts with pragmatic engineering to ship scalable ML systems.
10 years of coding experience
5 years of employment as a software developer
Baccalauréat, Mathématiques, mention très bien, Baccalauréat, Mathématiques, mention très bien at Lycee Moulay Youssef (Rabat, Morocco)Diplôme d'ingénieur, Data science, Diplôme d'ingénieur, Data science at Télécom SudParisexchange, Data science, exchange, Data science at Telecom ParisTech
English, Arabic, French, German
Github Skills (8)
pytorch10
machine-learning10
transformer10
nlp10
xnet10
python10
recommender-system10
testing9
Programming languages (2)
Github contributions (5)
NVIDIA-Merlin/Transformers4Rec
Apr 2021 - Jan 2023
Transformers4Rec is a flexible and efficient library for sequential and session-based recommendation and works with PyTorch.
Role in this project:
ML Engineer
ML EngineerContributions:93 reviews, 249 commits, 64 PRs in 1 year 9 months
Contributions summary:Sara contributed significantly to the `transformers4rec` library, primarily focusing on implementing and refining the integration of Transformer architectures for session-based recommendation. Their work involved adding and modifying features related to the Replacement Token Detection (RTD) and Causal Language Modeling (CLM) tasks within the XLNet architecture, alongside refactoring and commenting existing code related to these tasks. This included implementing new options, improving the processing of item interactions, and modifying the core meta-model architecture. The user also added testing code to validate implemented features.
xlnetbertlanguage-modelseq2seqrecommendation
sararb/Temporal-Deep-Learning-for-Medical-time-series
Jan 2022 - Feb 2023
Contributions:2 PRs, 1 branch in 1 year 1 month
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.