Akram Toumani

Training and coaching professionals and students in the fundamentals and advanced applications of AI: modeling, language processing, computer vision, and MLOps.

Development of concrete projects and workshops on responsible and explainable AI.

Keywords: Pedagogy · Python · TensorFlow · PyTorch · NLP · Computer Vision · AI Ethics · Knowledge Transfer

Creation and management of an applied R&D project aimed at designing a platform of conversational assistants augmented by human, emotional, and professional context.

Development of the Usine IA concept, an industrialization model for specialized conversational agents (health, education, coaching, HR…) based on RAG architectures, vector databases, and advanced conversational protocols.

Key achievements:

Design of the Sophia protocol, a contextual and non-therapeutic conversational framework, co-developed with a mental coach.

Structuring of a responsible AI framework balancing technical performance and emotional understanding.

Exploration of RAG models and ethical fine-tuning for optimal human-machine alignment.

Management of an AI innovation ecosystem around Sophia: testing, partnerships, monitoring, product validation.

Skills:

Generative AI · Machine Learning · Deep Learning · RAG · Product Design · Ethical AI Governance · Venture Building · Leadership · Strategic Communication

Design and execution of applied research projects in data science (CNRS, Research and Development Laboratory, HPM, etc.). Creation of big data pipelines (Python, SQL, PySpark). Training staff in data analysis and processing. Implementation of AI models (regression, classification, etc.) on real data.