From Harpy to LLMs: The Evolution of AI Learning
Explore the history of AI, from 1970s rule-based systems like Harpy to modern LLMs, and discover why moving beyond human imitation is the next frontier.
What is Grokking? The Geometry of LLM Intelligence
Discover how large language models learn through grokking and high-dimensional geometry. Explore the mathematical reality behind AI reasoning and generalization.
Small Language Models: The Sustainable Future of AI
Discover why Small Language Models are the efficient, eco-friendly alternative to energy-hungry AI, proving that bigger isn't always better for technology.
Andrej Karpathy — We're Summoning Ghosts, Not Building Animals
Andrej Karpathy on why LLMs are ghosts, not animals, and why education—not just AI—is the real bottleneck.
Evo: How Generative AI is Rewriting the Code of Life
Discover how Evo, a revolutionary generative model, treats DNA as language to design functional biological systems, shifting biology toward software engineering.
Beyond the UAT: Why Depth Beats Width in Neural Networks
Discover why depth outperforms width in neural networks. Explore the geometry of folding and how deep architectures solve complex problems better than shallow ones.