François Chollet
TL;DR François Chollet is a leading AI researcher, software engineer, and the creator of Keras, one of the most influential deep learning frameworks in the world.
François Chollet by Sora
François Chollet is a French computer scientist and artificial intelligence researcher best known for developing Keras, an open-source deep learning library that revolutionized how neural networks are built and tested. His work has made AI development more accessible, modular, and human-centered, enabling researchers, startups, and large companies alike to build complex models with ease.
Born in France, Chollet’s curiosity for intelligence, both artificial and human, has guided his career. After joining Google, he worked on TensorFlow integration and later contributed to the company’s AI and research divisions, focusing on deep learning, computer vision, and autonomous systems. Beyond engineering, Chollet is known for his intellectual depth and willingness to engage in philosophical discussions about the nature of intelligence, consciousness, and the limitations of current AI systems.
He is also the author of Deep Learning with Python, a seminal book that helped demystify neural networks for developers and students worldwide. Through his writings, talks, and open-source work, Chollet continues to advocate for human-aligned AI development, emphasizing robustness, ethics, and theory-driven understanding of intelligence rather than pure scale.
Created Keras, one of the world’s most popular deep learning frameworks.
Played a key role in integrating Keras with TensorFlow, shaping modern AI workflows.
Authored the influential book Deep Learning with Python, used by AI learners globally.
Contributed to major computer vision and machine learning projects at Google.
Developed the Abstraction and Reasoning Corpus (ARC), an AI benchmark exploring generalization and reasoning.
Advocated for human-centric, ethical, and interpretable AI development.
Recognized internationally as a thought leader in artificial general intelligence (AGI) discussions.
