Alex Kameni, ML Eng

Driven by a passion for data science and machine learning, I constantly seek opportunities to innovate, learn, and implement best practices. My collaborative spirit and commitment to continual improvement have fueled my growth in both research and applied domains.

About Me

I hold a Master’s degree in Complex Systems Engineering (2022), specializing in Machine Learning and Data Science from CY Cergy Paris University. My master’s research at the ETIS Laboratory of ENSEA explored data incremental learning for self-supervised computer vision models, addressing challenges in continual learning.

Additionally, I earned an engineering degree in Computer Science from the Polytechnic School of Yaoundé, Cameroon. My academic journey included in-depth research and two defended projects on cutting-edge topics in machine learning.

Currently, as a Data Scientist at Ivalua, I focus on enhancing invoice data capture and domain-specific language model integration. I’ve introduced transformative solutions like deploying LayoutLM for multimodal data comprehension, improving resource efficiency, and optimizing data pipelines. My contributions have significantly streamlined operations and advanced AI adoption within the organization.

Beyond work, I’m actively involved in projects spanning African history, NLP, computer vision, and self-supervised learning, showcasing my versatile skill set and commitment to impactful solutions.

Personal projects

Medivocate – Exploring African History and Culture with AI Published: January 11, 2025 An AI-powered platform exploring African history, culture, and traditional medicine, fostering understanding and appreciation of the continent’s rich heritage.

Fine-Tuning GLiNER for Location Mention Recognition (LMR) Published: September 01, 2024 Enhanced the GLiNER model to improve location mention recognition, aiding disaster response and location-based tasks.

Specializing Large Language Models for Telecom Applications Published: July 01, 2024 Enhanced Falcon 7.5B and Phi-2 models on telecom-specific knowledge using the TeleQnA dataset, leveraging Retrieval Augmented Generation and prompt engineering.

Object Detection Using Transformers Published: May 01, 2024 Developed a machine learning algorithm for accurate roof-type classification in rural Malawi using aerial imagery.

Semi-Supervised Learning with Few Labels Published: January 01, 2023 Improved self-supervised learning models in scenarios with minimal labeled data, enhancing their robustness and adaptability.

Continual Self-Supervised Learning Through Distillation and Replay Published: September 01, 2022 Addressed catastrophic forgetting in self-supervised learning using distillation, replay techniques, and predictive mechanisms.

Financial Data Generation Published: December 01, 2021 Developed methods for synthesizing high-quality financial data for machine learning applications.

Constraints Optimization of Resource Usage by Tasks in Workflows Published: September 01, 2021 Optimized resource allocation in workflows, balancing efficiency and performance.

NER for Commands Extraction Published: June 01, 2021 Designed a Named Entity Recognition system to extract commands from text, enabling better task automation.

Users’ Understanding Queries Published: September 01, 2020 Enhanced user query understanding by building NLP systems capable of semantic comprehension and accurate intent detection.