I am Youssef Assis , a Machine Learning Engineer with a PhD in Computer Science. My background spans both academic research and industry applications, which has honed my expertise in Software Engineering and Artificial Intelligence (AI). I am particularly interested in applying AI across diverse domains, focusing on Computer Vision (CV) and Natural Language Processing (NLP). Additionally, I have an interest in emerging areas such as Generative AI, tinyML, and explainable AI (XAI).
Software Engineering, AI, NLP
In 2018, I graduated with a bachelor's degree in computer science from the Faculty of Science and Technology (📍Beni Mellal, Morocco), ranking second in my class. During my time there, I worked on several projects, primarily in web and mobile applications. I then pursued a master's degree at the National Institute of Statistics and Applied Economics (📍Rabat, Morocco). This institution is recognized as a leading engineering school in Morocco and a frontrunner in the fields of statistics and AI. My master's program concentrated on Information and Intelligent Systems, integrating both theoretical and practical aspects, including Mathematical Modeling, Big Data, Image Processing, and Machine Learning (ML).
In March 2020, I moved to France for a six-month internship at the Engineering Science, Computer Science, and Imaging Laboratory (ICube) to explore new areas and challenges. Due to the COVID-19 pandemic 😷, my work was a combination of remote and on-site at the National Institute of Applied Sciences (INSA) in 📍Strasbourg, France. My project focused on developing a Question Answering (QA) system to analyze and extract knowledge from water and sanitation (RPQS) reports in PDF format.
This system leveraged the Bidirectional Encoder Representations from Transformers (BERT) model, a pioneering Transformer-based Large Language Model (LLM), and involved interaction with the HuggingFace library.
My project also incorporated Optical Character Recognition (OCR) techniques to extract data from images. The entire pipeline was integrated into a graphical user interface (GUI) to facilitate user interaction. Upon completing my internship and presenting my results, I earned my master's degree, achieving the second-highest rank in my class.
PhD, AI in healthcare, CV
My interest in AI and its applications in Healthcare led me to pursue a PhD focused on deep learning in radiology. In November 2020, I moved to 📍 Nancy, France, to work with the Tangram team at the National Institute for Research in Digital Science and Technology (INRIA).
My work focused on Computer-Assisted Detection (CAD) system, aiming to assist radiologists in their clinical practice. Specifically, the main goal of my research was to develop and evaluate an automatic system for detecting intracranial aneurysms from 3D Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) data. I was supervised by Dr. Erwan Kerrien and Prof. René Anxionnat. Additionally, I had the opportunity to collaborate closely with the diagnostic and interventional neuroradiology department at the Regional University Hospital (CHRU), a globally recognized leader in treating intracranial aneurysms.
I was involved in every stage of the project, starting with data annotation, then moving on to designing and training deep learning models, and finally evaluating their performance.
The experimentation involved various neural network architectures, primarily Convolutional Neural Networks (CNN), Attention Mechanisms, and the integration of a priori knowledge into the learning process.
Emphasizing a Data-centric approach, I prioritized the quality of training data over quantity, effectively tackling challenges arising from limited datasets and the class imbalance problem in medical data. This experience significantly enriched both my technical proficiency and personal growth, leading to significant scientific contributions, including the prestigious MICCAI STAR award 🏆. I successfully defended my PhD on March 22, 2024, with details available in my Defense slides, Manuscript, and the Examiners' report.
For a comprehensive view of my academic work, please refer to my publications listed on my Google Scholar page.
I am currently employed at DentalMonitoring (📍Paris, France) as a Research and Development (R&D) Engineer. In this role, I focus on improving a wide range of predictive models and applications used in the company's production environment. This involves evaluating the performance of existing models, identifying opportunities for improvement, and applying my expertise to enhance their performance.
Skills and interests
I have experience with a variety of technologies, tools, and libraries, as listed below. I am always eager to expand my knowledge and skills and am actively seeking new challenges where I can thrive in dynamic environments and address real-world problems. If you have a suitable position available, I would welcome the opportunity to connect and discuss how I can contribute to your team.
Throughout my career, I have worked with a diverse range of programming languages. Over the past year, my primary focus has been on Python, which I used extensively for most of my projects.
- Python
- C
- PHP
- Java
- JavaScript
- Matlab
- Bash/Shell scripting
- Latex
In my machine learning projects, I initially worked with TensorFlow for building and training models. However, in light of the increasing success and widespread adoption of PyTorch within the machine learning community, I transitioned to using it as my primary framework.
- PyTorch
- TensorFlow
- Scikit-learn
- OpenCV
- TensorBoard
- Wandb
I began with NLP projects, using RNNs, LSTM networks, and Transformers. I then transitioned to computer vision projects, exploring CNNs, attention mechanisms, and vision transformers.
- RNN / LSTM
- CNN
- GAN
- Attention mechanisms
- Transformers
In my work with medical imaging, I utilized DICOM and NIfTI formats for image storage and sharing and employed MONAI and TorchIO to develop deep learning models. I used ITK and SimpleITK for image processing, along with various visualization tools.
- DICOM / NIfTI
- MONAI
- TorchIO
- ITK / SimpleITK
- 3D Slicer
- ImageJ
- Nibabel
- MedPy
- Git / GitLab
- Poetry
- DVC
- Label Studio
- Qt
- Docker
- Linux
- JupyterLab
- GPU
- VSCode
- API REST
- Slack