{
  "id": 6291402,
  "title": "Prompt Engineering: How to Actually Get What You Want from AI",
  "url": "https://urgent.news/2026/09/08/prompt-engineering-how-to-actually-get-what-you-want-from-ai",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-08T15:31:56.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/blackwatch021/prompt-engineering-how-to-actually-get-what-you-want-from-ai-4e02"
  },
  "original_language": "en",
  "account": "Prompt engineering is the art of crafting precise instructions to elicit desired outputs from artificial intelligence. When you interact with an AI, the way you phrase your request can make a significant difference in the quality of the response you receive. For instance, if you ask an AI for interview preparation questions, the initial response may be generic and unhelpful. However, by specifying your role, relevant skills, and the desired format (such as a table), the AI can provide targeted information that is more useful to you.\n\nAt its core, prompt engineering involves structuring your prompts to provide the AI with the necessary context, information, and examples. This structured approach helps the AI understand your needs better, resulting in more accurate and relevant responses. One popular framework for crafting effective prompts is TCREI: Task, Context, References, Evaluate, and Iterate.\n\nThe \"Task\" component specifies what you want the AI to accomplish, such as generating a report, creating a tagline, or analyzing data. Adding a \"Persona\" to your prompt can change the tone and depth of the response by assigning a specific role or personality to the AI. For example, a software developer with 15 years of experience will provide different insights compared to a store manager with 10 years of experience.\n\n\"Format\" is another crucial element, as it defines the structure of the output you expect, such as a table, paragraphs, or CSV format. \"Context\" provides the AI with the necessary background information to better understand the task at hand. Providing relevant data or examples can significantly enhance the AI's ability to generate accurate and useful responses.\n\n\"References\" involve providing examples or source material that guide the AI in crafting its response. This can include showing examples of the desired output format, or supplying actual documents or data as the basis for the AI's analysis. The two main types of references are \"examples\" and \"source material.\" Examples help the AI understand the pattern or structure you want in the output, while source material allows the AI to generate accurate responses based on provided information.\n\nThere are different prompting techniques based on the number of examples provided: zero-shot, one-shot, and few-shot prompting. Zero-shot prompting relies solely on the AI's pre-trained knowledge, while one-shot prompting uses a single example to guide the AI's response. Few-shot prompting involves multiple examples, allowing the AI to identify patterns and generate more accurate outputs.",
  "summary": "You ask an AI to help you prepare for an interview. You get back twenty generic questions you could have found on any blog. You ask again. This time you mention that you are an entry-level developer, that you know the MERN stack, and that you want the answers in a table. Suddenly it is useful. Same model. Same knowledge. Different prompt. That gap, between the generic answer and the useful one,…",
  "key_points": [
    "Prompt engineering crafts precise instructions for desired AI outputs.",
    "TCREI framework structures prompts with Task, Context, References, Evaluate, Iterate.",
    "Zero-shot, one-shot, and few-shot prompting techniques vary in example usage."
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}