Foundations Readiness Checklist

Complete this checklist before proceeding to Level 1: Prompt Engineering. Minimum passing score: 10/12 items checked.

---

Mental Models

• I can explain what a "token" is and why token count matters for system design
• I can explain the difference between a context window and memory
• I can draw the Harness Architecture from memory (7 components)
• I can explain the 50/50 Rule and give an example of what belongs on each side
• I can name and describe at least 5 of the 10 LLM failure modes

System Design Readiness

• For my current project, I have defined the task boundary (where AI ends, determinism begins)
• For my current project, I have assessed the failure cost (what happens when the model is wrong)
• For my current project, I have defined the trust boundary (what the model can and cannot access)
• For my current project, I have an evaluation strategy (not "we'll check it manually")

Provider & Cost

• I understand the difference between API cost and total cost of ownership for AI features
• I have calculated a rough token budget for my primary use case
• I have a plan for provider abstraction (not hardcoding a single provider's API)

---

If You Cannot Check These Items

Missing Item	Read This
Mental models	01-how-llms-think.md
Harness Architecture	02-the-harness-principle.md
50/50 Rule	03-50-50-rule.md
Failure modes	04-failure-modes.md
Provider/cost	05-provider-landscape.md, 06-cost-mental-models.md
Latency	07-latency-tradeoffs.md