About me
I'm an AI research engineer and data scientist. At the end of 2021, I completed my Ph.D. in AI and joined Appvizer, a subsidiary of Softonic. As a digital marketing platform, Appvizer is dedicated to helping users find the most relevant software while enabling software vendors to reach their target audience. My research and development work at Appvizer primarily focuses on recommender systems and natural language processing. One of the most notable contributions has been the development of a hybrid recommender system, which led to an over 50% increase CTR (Click-Through Rate) compared to the previous legacy system. With the raise of LLMs in recent years, I have also been actively involved in prompt engineering and the development of innovative AI agents applications.
Past
After obtaining my Master's degree in Computer Science from Montpellier University in 2017, specializing in "Data, Knowledge & Natural Language Processing", I worked for Cirad for 1 year as a research engineer. Applying what I had learnt during my Master's studies, I developed a tool based on Semantic Web to facilitate information retrieval, data analysis and visualisation. Although the tool proved useful and was well-received by the firm's agents, I realised that I need to deepen my knowledge and further refine my problem-solving and analytical skills.
To this end, I chose to pursue a Ph.D. in AI---a challenging yet deeply rewarding achievement for me. During my Ph.D. at the CERIS laboratory within IMT Mines Alès, I focused on recommendation systems, a key application of machine learning widely used in real-world scenarios to mitigate information overload for end-users. The primary goal of my thesis was to enhance recommendation performance from a data-to-knowledge perspective. The data perspective aimed to improve recommendation accuracy (i.e., rating prediction and ranking) by leveraging statistical models. Meanwhile, the knowledge perspective went a step further, focusing on enhancing the quality of recommendations through more user-centric aspects such as diversity and explainability. This was achieved by incorporating knowledge engineering concepts, including the Semantic Web and knowledge graphs.
Training-wise, I have a comprehensive background in both machine learning and knowledge engineering. My Ph.D. research has helped me refine my technical skills, making me more proficient with major machine learning frameworks, containerization, and other relevant technologies. More importantly, it has also allowed me to develop essential professional skills such as presentation, writing, critical thinking, and teamwork.
Outside of work, I enjoy cooking delicious meals, travaling, reading, reflecting, and sharing life experiences with friends---especially my wife 👸. I'm also passionate about self-learning, particularly in emerging technologies.
Domains of interest
- Machine learning
- Prompt engineering
- AI agents
- Recommender systems
- Knowledge engineering
- Knowledge graphs
- Information retrieval
- Data analysis