Sub-Track
Fine-Tuning and Adaptation
Adapt real models with LoRA/QLoRA on a student GPU budget.
The adaptation lifecycle, full vs parameter-efficient fine-tuning and the memory math, LoRA and QLoRA hands-on, instruction tuning and preference optimization (DPO/RLHF), data curation and decontamination, and the compute realities of free/cheap GPUs. Ends with a measured fine-tuning experiment and a job-readiness lesson.
8 lessonsIntermediate → Advanced~10h total
Prerequisite Sub-Track
Transformers and NLP Foundations →Complete this sub-track before starting Fine-Tuning and Adaptation.