Prompt Engineering

Five ideas that change how you build with AI.

Not a tutorial. A set of lenses — architectural decisions that determine whether a system behaves reliably or just seems to work.

4 concepts · viewing 1 of 5

01

Layering

Your prompt is a stack of sources with different authority and lifespan — not one blob of text.

Design choices
  1. 01
    Precedence
    What wins when two layers conflict — a user instruction versus a retrieved document, a system rule versus a session preference.
  2. 02
    Provenance
    Whether the model can say where a claim came from — which layer it lives in, so contradictions can be traced back to their source.
  3. 03
    Budget allocation
    Which layer gets cut first when the context window fills — and whether that decision is made deliberately or left to chance.
  4. 04
    Layer ordering
    Whether the sequence of layers in the prompt actively signals relative importance, not just logical grouping.
  5. 05
    Refresh cadence
    Which layers are fixed for the session and which are rebuilt per turn — and whether stale content persists past its useful life.
Typical root causes
  1. 01
    Instructions ignored after a long conversation
    The task was stated once, near the top, and gets deprioritized as later turns fill the window — the "valley of meh." The model never forgot; the instruction just lost effective weight.
  2. 02
    The model contradicts something said early on
    A later layer — a retrieved document, a follow-up message — implicitly overrides an earlier constraint. No conflict resolution rule exists, so recency wins by default.
  3. 03
    Retrieved content overrides an explicit user rule
    Retrieval results appear after the user instruction in the prompt, so the model weights them as more recent and therefore more authoritative.
  4. 04
    Can't tell a user's rule from a random file's suggestion
    Without provenance markers, the model has no basis for deciding which source carries more weight — everything looks equally authoritative.
  5. 05
    Behavior changes just from reordering the same content
    The prompt structure isn't encoding importance intentionally, so the model infers it from position — and that inference is fragile.
Context window34% used
System
always present
Developer
Tools
Retrieved context
History
Memory
Task, restated
always present
Recognize this?
1 / 5