HuggingFace LLM Course

HuggingFace's most popular LLM course with 50k+ GitHub stars. Two tracks: LLM Scientist (architecture, pre-training, fine-tuning, quantization, evaluation) and LLM Engineer (running, RAG, deployment, app development). Uses core HuggingFace libraries: Transformers, Datasets, Tokenizers, Accelerate.

For developers and researchers wanting to systematically understand LLMs. Scientist track for those diving into model internals, Engineer track for those building applications quickly. Basic Python and deep learning knowledge required.