Benezra Raphael

Hello, I am a seasoned Data Scientist, Machine Learning and Computer Vision Engineer with a track record of devising and implementing advanced algorithms to solve complex problems. My academic background includes a Master's degree in Computer Science, followed by extensive professional experience across various industries.

Previously, I've served as a Data Scientist at Saint-Gobain and a Computer Vision Engineer at Securaxis. In my most recent role as a Machine Learning Engineer at Kili Technology, I've further honed my skills in building and deploying machine learning models.

My technical expertise encompasses programming languages like Python and C++, alongside proficiency in libraries and frameworks such as OpenCV, TensorFlow, and PyTorch. My passion lies in staying at the forefront of data science and computer vision research, and integrating cutting-edge techniques into existing systems.

On a personal note, I am a detail-oriented, self-motivated professional who takes a methodical approach to problem-solving. I'm known for my strong interpersonal skills and have a knack for leading and mentoring junior team members.

I'm currently exploring new opportunities where I can apply my skills and contribute to innovative projects in the field of data science and machine learning. I look forward to connecting with potential employers and professionals in the industry.

Senior Data-Scientist at Saint-Gobain Distribution Bâtiment France (October 2021 - Present, Paris, France)

At Saint-Gobain, I have been engaged in several complex projects, utilizing techniques like natural language processing for thematic extraction from client feedback. By implementing TF-IDF for word representation and LDA for topic detection, I was able to construct a robust data processing pipeline using Scikit-learn and NLTK, which was deployed on a cloud platform via Docker and Kubernetes. This greatly expedited the identification of recurrent issues, enabling the company to proactively target improvement efforts.

Another notable project was the design of a business potential model for La Plateforme Du Bâtiment. I utilized client visit trends to drive commercial actions, resulting in a successful regional deployment. I also developed a consumption embedding model for the client base using an auto-encoder which significantly enhanced performance in multiple productionized machine learning models like potentials, churn, segmentation, and more.

I also computed departmental market shares using linear models on INSEE data. This provided demonstrable results on proxy patterns, as direct measurements were impossible. Moreover, I implemented a client segmentation model for DSC with a feedback-loop labeling system for continuous learning.

Tech Stack: Python, Tensorflow, Pytorch, Spark, Docker, Kubernetes, Azure, Databricks, SQL, MongoDB.

Data Scientist at SecuraXis (January 2020 - October 2021, Geneva, Switzerland)

In SecuraXis, I was responsible for engineering Python-based data science programs for sound analysis and vehicle classification. The algorithms I developed were deployed on connected acoustic sensors in various European cities. I leveraged vision models and CNNs applied to mel-spectrograms and GCC-PHAT to achieve accuracy rates exceeding 90%.

I developed a specific model for bird species inventory based on mel-spectrogram analysis, which was successfully deployed in several natural parks across France. I also built a comprehensive data science pipeline for sound-based analytics, including vehicle detection, classification, and bird species identification, incorporating a robust testing and monitoring system to ensure optimal performance of productionized models.

Tech Stack: Python, Tensorflow, OpenCV, Pytorch, Keras, Jupyter, GPU computing, Docker, Flask.

Junior Data Scientist at Rakuten Advertising (March 2019 - December 2019, Paris, France)

During my tenure at Rakuten Advertising, I led an initiative to create effective betting strategies in the ad-exchange business, which required advanced mathematical and analytical skills. I implemented an RNN-GRU model for product recommendation, resulting in over a 10% improvement in click-through rates as measured through AB-testing.

Tech Stack: Python, Keras, TensorFlow, Spark, MongoDB, SQL, GCP.

I am a passionate machine learning engineer, dedicated to utilizing my skills for impactful projects. Throughout my career, I have demonstrated expertise in developing machine learning projects, auditing data infrastructures, and training data-focused teams. My reliability and strong communication skills have been an asset in all my professional roles.

Engineering Diploma - MSc from Centrale Lyon (September 2015 - April 2019, Lyon, France): I majored in mathematics and data science during my time at this esteemed institution, ranking among the top 10% with an overall GPA of 3.73/4.

Machine Learning and Deep Learning Specialization by Andrew NG - Stanford (2019): This course deepened my understanding of machine learning and deep learning, imparting me with the skills to construct robust models and gain valuable insights from complex data sets.

How Google does Machine Learning & Launching into Machine Learning by Google (2019): These courses provided me with insights into how one of the world's tech leaders, Google, approaches machine learning, enriching my understanding of the field and exposing me to new perspectives and methodologies.

My education before these was a period of rigorous scientific and mathematical training at the prestigious Lycée Thiers, where I pursued MSPI/MP* (September 2012 - June 2015, Marseille, France). This period served as a strong foundation for my subsequent studies and career in data science and machine learning.