Welcome to my digital lab! I spend my days trying to teach Neural Networks how to follow the laws of logic, and my nights wondering why they still prefer to hallucinate.

I operate at the intersection of two worlds that I am determined to merge:

• The Researcher (PhD): My work focuses on Neuro-Symbolic AI. I’m on a mission to marry the statistical power of Deep Learning with the explainable rigor of Symbolic Reasoning. Basically, I want AI to stop just predicting and start reasoning.
• The Consultant (@ CGI): I help organizations navigate the AI hype. I turn "Generative AI" buzzwords into robust, scalable, and—most importantly—reliable industrial solutions.

My Philosophy: If a model needs 10 billion parameters to realize a cat isn't a dog, it might be missing a bit of common sense (and a few logical rules).

• Deep Learning: PyTorch (my comfort zone), TensorFlow.
• Symbolic AI: Logic Programming, Knowledge Graphs, Ontologies, formal reasoning.
• Generative AI: LLMs, RAG Architectures, Prompt Engineering, and fine-tuning.
• The Classics: Python, Scikit-learn, and the occasional SQL query when reality hits.

• Explainability: Peeking inside the "Black Box" to see what's actually happening.
• Logic Injection: Forcing neural nets to respect domain knowledge and physical constraints.
• AI Ethics: Because a "smart" AI is good, but a "responsible" AI is non-negotiable.

As you might have noticed, my avatar was generated by Nano Banana.

The Challenge: Can you guess the prompt? Hint: It involves a tropical drink, some complex math, and the idea that even an AI researcher dreams of the beach (or perhaps the idea that the AI works while I sip a cocktail... I'll let you interpret the latent space).

Whether you want to discuss First-Order Logic, the challenges of deploying LLMs in production, or debate if a Transformer truly "understands" what it's saying:

• LinkedIn: Elie El Debs
• Email: elie.eldebs@outlook.fr

"In a world of gradient descent, stay symbolic." 📐✨