LLM Fundamentals
Transformer architecture, attention, tokenization, inference, and sampling strategies
Prompting Techniques
Zero-shot, few-shot, chain-of-thought, prompt templates, structured output, and prompt chaining
Embeddings & Vector Stores
Embedding models, similarity search, vector databases, and indexing strategies
RAG Basics
Retrieval-Augmented Generation pipeline, chunking strategies, retrieval methods, and evaluation
Fine-Tuning Overview
When to fine-tune, LoRA, QLoRA, RLHF, DPO, and decision frameworks for choosing the right approach
Evaluation Metrics
Traditional metrics, LLM-as-judge, human evaluation, RAG evaluation, and agent evaluation approaches