Neural networks are the technology that power modern large language models. But how does that black box actually work? What does training a model really do? This technology has become a daily tool for many people, so this session focuses on understanding the ideas behind it.

In this session we'll build our own small and intentionally simple implementation of a neural network in C#. Aimed at developers, we'll keep the math to an approachable minimum and concentrate on the underlying algorithms. We'll implement and understand neurons, weights, and biases. From there we'll connect these fundamentals to the transformer architecture and how it enables modern large language models.

After this session you'll have a clearer understanding of the black box behind neural networks. You'll see the impact of weights and biases by building an implementation yourself in a familiar language, and you'll walk away with a practical mental model of how today's AI systems operate.