
A clean, theory-first path from classic deep learning to today’s generative models and a practical finish with fine-tuning. In three focused sessions, you’ll build solid intuition for MLP/CNN/RNN, decode modern families (Transformers, Diffusion, SSMs/Mamba, MoE), and wrap with a 20-minute live code sprint that fine-tunes a small text model. Slides, tiny snippets, and clear trade-off thinking no heavy math, no big projects.