The neuron analogy · Layers and weights · Activation functions · Training (gradient descent intuition) · CNNs vs. RNNs vs. Transformers · GPU economics · When deep learning is worth it