How modern LLMs work, end to end — attention and transformers, building a GPT, efficient inference, fine-tuning with LoRA/RLHF, and retrieval & agents. Interactive, every number real.
Who it's for
Engineers and learners who want to understand — and build — modern language models from the ground up.
What you'll learn
Understand attention and the Transformer architecture
Follow next-token prediction and text generation end to end
Reason about efficient inference — KV-cache, speculative decoding, quantization
Grasp fine-tuning and alignment — LoRA, RLHF, RAG, and agents