The skill of crafting and refining instructions for AI language models to produce accurate, relevant, and high-quality outputs across diverse tasks.
Prompt engineering is the practice of designing effective inputs for large language models. It spans from writing clear queries to architecting multi-step workflows with chain-of-thought reasoning, system prompts, and agent orchestration, bridging human intent and machine capability.
You have begun interacting with AI chatbots and can obtain simple answers. You accept outputs at face value without evaluating quality or iterating. You are learning that phrasing directly affects the response, and you are starting to explore what these tools can and cannot do.
What Comes Next
If you've checked off most of this list, you're ready for the Beginner, Structuring stage, structuring prompts with deliberate context, roles, and formatting instructions. Bandura(1977)'s Social Learning theory suggests watching prompt crafting demonstrations and studying AI output examples builds the confidence to write structured prompts on your own.
A 5-level prompt competency framework (Beginner-Expert) defining behavior-based proficiency boundaries, used to calibrate checklist difficulty across levels.
Systematic learning path from basic prompting to agent design, providing criteria for progressive technique difficulty and advanced technique classification.
Systematic learning path covering clear instructions, example provision, XML structuring, role prompting, thought elicitation, and prompt chaining, used as technical evidence for L2-L4 checklist items.
Empirical evidence that chain-of-thought prompting elicits reasoning in large language models, providing academic authority for the L3 CoT checklist item and the overall technique difficulty hierarchy.
