Barbados Traffic Analysis Challenge
ML solution for predicting traffic congestion using video analysis. Implemented YOLO detection, VideoMAE embeddings, and ensemble models.
Wendyellé A. Alban NYANTUDRE
Machine Learning Engineer
Safi, Morocco · he/him
« Ad astra, per aspera »
Most people just call me Alban. My last name NYANTUDRE means hope, or more fully: as long as there is life, there is hope.
I’m an engineer. In July 2026 I graduated with a Diplôme d’Ingénieur d’État in Computer Science and Artificial Intelligence from the École Nationale des Sciences Appliquées in Safi (Cadi Ayyad University).The routeThe Moroccan grande école path is a competitive and prestigious route into engineering: two years of classes préparatoires, then three years of engineering, with no degree awarded in between, ending in the Diplôme d’Ingénieur d’État. I have strong foundations in applied mathematics and computer science with a passion for machine and deep learning. I’m currently open to PhD positions. My research interest is broadly deep learning, and more particularly NLP, computer vision, multimodal learning, multilingual models, speech technologies and reinforcement learning.
Alongside my previous studies, I contributed to training and deploying translation, speech recognition and speech synthesis models for Mooré, a low-resource language with over 7 million speakers and very little data.The 5-LOX workIn this work we trained a hybrid quantum-classical variational autoencoder that generates small molecules able to act on 5-lipoxygenase (5-LOX), an enzyme involved in many inflammatory diseases. Several of the molecules it generated showed better binding affinity and drug-like properties than zileuton, the only approved direct inhibitor. Most recently, from March to August 2026, as a Mitacs Globalink research intern, I worked on hybrid quantum-classical generative modelling for the de novo design of 5-lipoxygenase inhibitors (for the treatment of inflammatory diseases) at the Université de Moncton (Canada).
Feel free to explore my projects, check out my blog, or get in touch.
Grade: Highest Honours, 17.25 / 20 · Rank: 1st / 19.
Coursework: applied mathematics, data structures & algorithms, stochastic modelling, optimisation, NLP, heuristics, deep learning, data mining, big data, computer vision, game theory, cloud computing, fuzzy systems.
Grade: Highest Honours, 16.25 / 20 · Rank: 3rd / 262.
Coursework: calculus, probability & statistics, linear algebra, algorithms, data structures, C programming, electronics & computer architecture, physics.
Collected a dataset of 5-lipoxygenase (5-LOX) inhibitors from chemical databases (ChEMBL, PubChem, BindingDB) and implemented hybrid quantum-classical generative architectures (QGAN, QVAE) for the de novo generation of 5-LOX inhibitors.
Designed, trained and deployed translation, automatic speech recognition (ASR) and text-to-speech (TTS) models for Mooré, a low-resource West African language with over seven million speakers and the most widely spoken language in Burkina Faso.
Developed and deployed ML models to predict prepayment risk of mortgage-backed securities, supporting better risk assessment for financial institutions.
ML solution for predicting traffic congestion using video analysis. Implemented YOLO detection, VideoMAE embeddings, and ensemble models.
Python project for extracting and processing data from Spotify Web API with OAuth 2.0 authentication and pagination strategies.
Open-source toolkit to collect, preprocess, align, and normalize Mooré language speech/text data for low-resource NLP applications.
ETL pipelines and Deep Learning models to forecast droughts from satellite data, helping Moroccan farmers optimize water resources.
Details2025