Machine Learning Engineer, Computer Vision
2021 - PRESENTBusPatrol- Contributed to several computer vision projects using state-of-the-art CNN and vision transformers to solve a boundary detection problem with over 98% accuracy and optimized the neural networks to double the speed gain on CPU inference.
- Led the deployment of a machine learning (ML) lifecycle management infrastructure on AWS, increasing efficiency and reproducibility of ML workflows internally and easing collaboration with the DevOps team.
- Addressed a business-critical problem, which was first thought to need AI, through simple data aggregation and analysis. The surfaced insights brought a simple way to solve the issue, saving the company significant time and cost.
- Deployed production-grade ML Inference Infrastructure to AWS as microservices using AWS CDK and AWS SageMaker Endpoints.
Technologies: PyTorch, Python, Docker, OpenCV, SciPy, Machine Learning, Artificial Intelligence (AI), Machine Learning Operations (MLOps), Amazon Web Services (AWS), You Only Look Once (YOLO), NumPy, Deep Learning, Computer Vision, Image Processing, Scikit-image, Scikit-learn, TensorFlow, Data Science, Convolutional Neural Networks, Amazon SageMaker, Infrastructure as Code (IaC), Pandas, 3D Image Processing, Object DetectionMachine Learning Engineer, Computer Vision
2020 - 2021Faimdata- Developed the company's first optical character recognition algorithms and pipeline in Python, C/C++, PyTorch, TensorRT, and DeepStream, succeeding with 95% accuracy in the first pilot project, leading to the company acquiring the client's business.
- Applied transfer learning from the first deployments to a new use-case, reproducing similar performance levels, succeeding in a second pilot project that led to the acquisition of a second client's business.
- Reduced iteration time by 25% after establishing workflow best practices in ML deployment iterations using Docker, Google Cloud Platform, Git, Data Version Control, and model testing.
- Pruned deep neural networks (DNNs) to 60% faster runtime, allowing for the use of more affordable hardware and reducing hardware expenses by 15%.
- Hired and trained a new team member and transferred knowledge to help grow and scale the company's computer vision team.
Technologies: TensorRT, NVIDIA Jetson, OpenCV, PyTorch, Python, C++, CUDA, CMake, Google Cloud Platform (GCP), Node.js, Docker, DeepStream SDK, Artificial Intelligence (AI), Swift, You Only Look Once (YOLO), NumPy, Deep Learning, Computer Vision, Image Processing, Machine Learning, SciPy, Scikit-image, Scikit-learn, TensorFlow, Data Science, Convolutional Neural Networks, Pandas, 3D Image Processing, Object Detection
Reda Oulbacha
Artificial Intelligence Developer in Montreal, Canada
Member since August 9, 2022
Reda earned his M.Sc doing Computer Vision Machine Learning research at the University of Montréal, École Polytechnique. He published at IEEE and Wiley, and published a PCT patent application. He then built real-world Computer Vision AI systems in the logistics and transportation safety industries. His previous projects include a PyTorch framework for medical images, a CV method for 3D CT to 2D X-ray image fusion, and a kernel pruning method for YOLOv5. Reda loves working on innovative products.
Reda is now available for hire
Experience
- Artificial Intelligence (AI) 4 years
- PyTorch 4 years
- Deep Learning 4 years
- Computer Vision 4 years
- Python 4 years
- Scikit-learn 3 years
- Machine Learning 3 years
- TensorFlow 2 years
Availability
Part-time
Preferred Environment
C++, PyTorch, TensorFlow, OpenCV, CUDA, Scikit-learn, Scikit-image, ARKit, Python, Data Science
The most amazing...
...thing I've worked on is a method for MRI-guided spine surgery using CycleGAN with very limited data, which we published in a peer-reviewed scientific journal.
Employment
Experience
- Development of a PyTorch Framework for Medical Imageshttps://github.com/Roulbac/GanSeg
A framework for neural networks in PyTorch, with easy tools to process the most common file formats for medical images and support for model checkpointing, model export, model retraining, and a visualization board implemented with the Vizdom framework.
- A Computer Vision Tool for 3D CT to 2D X-ray Image Fusionhttps://github.com/Roulbac/2D3DAutoReg
A computer vision tool that I developed in Python using NumPy, SciPy, Numba, and CUDA to estimate the fusion transformation between 3D CT scans and 2D X-rays. This procedure is commonly used in medical image computer vision research, a problem that this tool solves.
- Integration of a Structured Pruning Method for YOLOv5https://github.com/Roulbac/yolov5/tree/feature/torch_pruning_integration
YOLOv5 is a commonly used object-detection DNN. This project dissected the neural network and divided a strategy to allow the pruning and retraining of any YOLOv5 neural network. The outcome was a 60% average increase in inference speed and a model size reduced to only a few megabytes.
- ARKit iOS Application to Denoise the Camera Pose Using CoreML on a DJI Gimbal
An application using CoreML, PyTorch, and ARKit to denoise the ARKit camera pose tracking when mounted on a DJI handheld gimbal. The system consists of a pre-designed PyTorch model in Python that is exported on the iPhone under CoreML and re-trained with runtime data.
Skills
Languages
Python, C++, SwiftLibraries/APIs
PyTorch, TensorFlow, OpenCV, Scikit-learn, SciPy, NumPy, Node.js, PandasTools
Scikit-image, You Only Look Once (YOLO), CMake, Amazon SageMaker, ITK, Xcode, DJI SDKParadigms
Data ScienceOther
TensorRT, NVIDIA Jetson, DeepStream SDK, Machine Learning, Image Processing, Computer Vision, Deep Learning, Numba, Artificial Intelligence (AI), Convolutional Neural Networks, Object Detection, Generative Adversarial Networks (GANs), Medical Imaging, Machine Learning Operations (MLOps), Sequence Models, Infrastructure as Code (IaC), 3D Image Processing, Augmented Reality (AR), Time Series, Natural Language Processing (NLP)Frameworks
Core ML, ARKitPlatforms
Google Cloud Platform (GCP), Docker, Amazon Web Services (AWS), CUDA, iOS
Education
- Master's Degree in Biomedical Engineering2017 - 2019École Polytechnique (Affiliated with University of Montréal) - Montréal, Québec, Canada
- Master's Degree in Electrical Engineering2013 - 2019INSA Lyon - Lyon, France
Certifications
- DeepLearning.AI TensorFlow Developer Professional CertificateFEBRUARY 2020 - PRESENTCoursera
- Deep Learning SpecializationDECEMBER 2018 - PRESENTCoursera
