Prompting & Context
Everything you do at inference time to get more out of a model. The structure of a good prompt, the surprises of in-context learning, and the failure modes that emerge when context gets long.
Prompting Fundamentals
The repeatable structure of a prompt that works — system, instruction, examples, constraint.
In-Context Learning
Teaching a model a task using only examples in the prompt — and why it works at all.
Chain-of-Thought & Reasoning
Asking the model to think out loud, and why it improves accuracy on hard problems.
Prompt Caching
Reusing the KV state for prefixes you send repeatedly — the latency and cost win.
Context Engineering
Designing the whole input — system prompt, history, tools, retrieved docs — as one coherent system.
Context Pathologies
Rot, drift, lost-in-the-middle — and the limits of long-context windows in practice.
Prompt Injection & Jailbreaks
The security surface where untrusted text meets a trusting model.