Jean-Marc Beaujour is a Senior Data Scientist at BrightApps, responsible for model development and implementation, data analytics and insights, collaboration and communication with client-side software developers and stakeholders, and driving innovation and improvements based on the latest AI advancements. He holds a Ph.D. in physics, bringing 15 years of experience working in R&D environments for startups and Fortune 500 companies.
Recently, Jean-Marc developed a comprehensive pipeline to optimize open-source LLMs aimed at providing natural language explainability for a recommendation engine. This project improved the transparency of the generated recommendations and enhanced user trust and satisfaction in the system. Additionally, he developed a multi-agent framework using LLMs to identify potential clients, streamline the client acquisition process, and uncover new business opportunities.
Jean-Marc’s expertise lies in solving real-world problems using machine learning and deep learning. With a strong background in physics, he has the ability to analyze and evaluate problems at a high level and apply cutting-edge technologies to support his expertise in machine learning and AI. His passion for creating innovative solutions and his talent for integrating advanced AI techniques make him a valuable asset to the BrightApps team.
Programming
Python
PyTorch
TensorFlow
Innovation Solutions
Creates innovative cutting edge solutions for government and business through Internet of Things, Edge computing for Deep Learning models and realtime monitoring (sensors).
Supervised/Unsupervised Machine Learning
Linear/Logistic Regression
Support Vector Machines
Random Forest
Naïve Bayes Classification
Recommender systems
Ensemble Methods
Clustering algorithms
Principal Component Analysis
Deep Learning
Object Detection
Image Classification
Image Segmentation
Model for Sequence Data (RNN, LSTM)
Deep Reinforcement Learning
Computer Vision
Affine Transform
Homography
Visual Perception
Feature Extraction
Reinforcement Learning
Model-based/Model-Free Reinforcement Learning
Deep Reinforcement Learning
