Alouini Yassine
Verified Expert in Engineering
Machine Learning Developer
Metz, France
Toptal member since May 19, 2022
Alouini is a seasoned data scientist with over seven years of experience building internal tools and products with ML engineering and applications for computer vision tasks like OCR, object segmentation, and depth estimation. He enjoys contributing to projects where he can explore some of its R&D aspects, implement a robust model, build a quick prototype to show how it works, and finally deploy it to production.
Portfolio
Experience
- Python - 8 years
- Machine Learning - 7 years
- Pandas - 7 years
- Computer Vision - 5 years
- Deep Learning - 4 years
- PyTorch - 3 years
- Visual Studio Code (VS Code) - 3 years
- Streamlit - 2 years
Availability
Preferred Environment
PyTorch, Visual Studio Code (VS Code), Streamlit, SQL, Artificial Intelligence (AI)
The most amazing...
...ML model deployments I've performed made them more reliable and quicker, down from many hours and manual processes to 30 minutes and semi-automated.
Work Experience
Senior Data Scientist | Computer Vision Engineer
Fieldbox
- Explored various computer vision tasks, like OCR, segmentation, and counting, and applied them to an industrial setting, which helped the pre-sales team win at least three computer vision projects.
- Helped start the machine learning engineering team by rewriting existing tools into a more modular and modern suite of Python packages. These tools are mainly used to ease the database and file systems connection.
- Developed and deployed five computer vision POC models to staging servers and used the resulting API to develop visualizations.
- Implemented a semi-automatic labeling workflow for instance segmentation using CVAT (computer vision tool) and SAM (segment anything model).
Data Scientist
Qucit
- Worked on urban mobility data analysis using mainly geographical and contextual data. I also performed visualizations and consultancy to help the sales team tailor their response to the client's need, mainly in the public mobility sector.
- Implemented mobility predictive models—like bike-sharing, parking, and fraud—used in many B2B products to help operators understand their system's usage and optimize it.
- Launched a series of internal talks that improved overall data science and engineering expertise and gave a technical meetup presentation that helped promote the company's work and attracted a few candidates to apply to data science jobs.
Experience
Top 4% on the Kaggle Champs Competition
https://www.kaggle.com/c/champs-scalar-couplingThe model uses the molecules' XYZ structure and some of the atoms' chemical properties. This model could be helpful to speed up the necessary chemistry computations, including quantum properties, to be exact.
Various Deep Learning And Computer Vision Notebooks
Some of the notable ones include:
1. A notebook exploring segmentation metrics: https://www.kaggle.com/yassinealouini/all-the-segmentation-metrics
2. A notebook explaining how the RoBERTA model runs and how to train it on TPU: https://www.kaggle.com/yassinealouini/roberta-meets-tpus
3. A notebook explaining how EfficientDet works and how to train one using PyTorch Lightning: https://www.kaggle.com/yassinealouini/efficientdet-meets-pytorch-lightning
Education
Master's Degree in Statistics
University of Cambridge - Cambridge, UK
Master's Degree in Engineering
Ecole Centrale Paris - Paris, France
Skills
Libraries/APIs
PyTorch, Pandas, NumPy, Scikit-learn
Tools
GitLab CI/CD
Languages
Python, SQL
Platforms
Visual Studio Code (VS Code), Jupyter Notebook
Frameworks
Streamlit, Flask
Storage
PostgreSQL
Other
Computer Vision, Artificial Intelligence (AI), Data Science, Machine Learning, Deep Learning, Detectron2, Optical Character Recognition (OCR), Data Engineering, Data Analytics, Back-end Development, Statistics, Applied Mathematics, Probability Theory, Engineering
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring