Learning objectives
- • Explain self-attention, positional encoding, and multi-head attention
- • Contrast encoder-decoder setups with decoder-only LLMs
- • Reason about context length, memory, and inference cost
Deep Learning
Transformers From First Principles
Understand the transformer stack deeply enough to explain scaling, context handling, and attention trade-offs.
Recommended on day 30120 minutesAdvanced
Interview prompts
- • Why do transformers parallelize better than RNNs?
- • What breaks when context windows grow without retrieval support?
Prerequisites