My Blog
Notes on AI engineering, LLMs and building production systems
Transformer Architecture Explained: How Transformers Work
Transformer architecture explained step by step: embeddings, attention, feed-forward layers, residuals, causal masks and why generation is sequential.
How Large Language Models Work: A Practical Guide
Learn how large language models work: tokens, next-token prediction, temperature, pretraining vs fine-tuning, and why LLMs hallucinate.