What is Prompt engineering?
Also called: Prompting · Prompt design · Prompt writingDefinition
Prompt engineering is the practice of designing the instructions and context given to a generative AI model so that it produces reliable, useful output. It involves stating the task and goal clearly, supplying relevant context and source material, specifying the format and audience, giving examples, setting constraints, and testing and refining the prompt against real cases. For most professionals, it is less about tricks and more about clear briefing, the same skill needed to delegate work well to a capable colleague.
A strong prompt typically includes the role or perspective the model should take, the task, the background it needs, the material to work from, the output format, and what to avoid. Asking the model to work through steps, providing one or two worked examples, and breaking a large task into smaller prompts all tend to improve results.
At work, prompt skill is what separates a generic first draft from output that is genuinely usable. An HR professional who supplies the role context, competencies and company tone gets a far better job description than one who types a single line. In software, prompts become part of the application and are versioned and tested like code.
Common mistakes are vague one-line requests, leaving out the source material and then trusting answers the model had to guess, and never checking the output. Another is believing a perfect prompt removes the need for review; prompting improves quality but does not guarantee accuracy.
Key points
- State the task, goal, audience and format clearly.
- Provide context and source material, not just a request.
- Use examples and break large tasks into steps.
- Test and refine against real cases.
- Always review output; prompting does not guarantee accuracy.
An example at work
An HR business partner in Gurugram prompts an approved AI tool with the role’s responsibilities, the company’s competency definitions and three past interview questions, asking for a structured interview guide with scoring anchors.
Where this is used at Bodhih
Related terms
Generative AI
Generative AI is a type of artificial intelligence that creates new content, such as text, images, audio, video or code, from patterns learnt from data.
Large language model
A large language model (LLM) is an AI model trained on vast amounts of text to understand and generate language by predicting the next token.
AI literacy
AI literacy is the ability to understand what AI can and cannot do, use it effectively and safely, and judge the quality of its output.
AI hallucination
An AI hallucination is output from a generative AI model that sounds confident and plausible but is false, unsupported or made up.
LLM evaluation
LLM evaluation is the systematic testing of a language model or LLM application against defined criteria to measure quality, accuracy and safety.
More about Prompt engineering
Is prompt engineering still a skill worth learning?
Yes, though it has shifted. Models now handle loose requests better, so tricks matter less. What still matters is clear briefing: supplying the right context, defining what good output looks like, splitting work into steps and checking results. For developers, designing and testing prompts inside applications remains a core part of LLM engineering.
What makes a good prompt?
A good prompt tells the model what to do, why, for whom, using what material, in what format, and what to avoid. It is specific about the output and includes relevant context. Adding an example of the result you want and asking the model to flag uncertainty usually improves quality further.