Prompt Engineering

Prompt Engineering Fundamentals

Learn the building blocks of a great prompt — role, context, task, format, and constraints — and how ChatGPT, Claude, Perplexity, and Grok each respond differently to the same instructions.

Build a Personal Study Assistant

Turn a stack of lecture notes into a study partner that quizzes you, explains what you got wrong, and tracks which topics you keep failing. You will finish with a reusable system prompt you can point at any subject.

Run an Honest CV Review

Build a review process that tells you what is weak in your CV instead of politely rewriting it. Includes a screening simulation that shows how it reads in the six seconds it actually gets.

The Hostile Interview

Get a model to interview you properly — following up on weak answers instead of moving politely on. Uncomfortable, and considerably more useful than rehearsing answers you already have.

Summarise Without Losing the Point

Compress a long document to a tenth of its length while keeping what matters, then verify nothing important was dropped. Tests whether you can direct compression rather than just requesting it.

What Is Prompt Engineering? A Practical Guide for 2026

Prompt engineering is the craft of writing model inputs that reliably produce the output you want. This guide covers the five-part framework, the techniques that actually matter, and the mistakes that make output generic.

10 Prompt Mistakes That Make AI Output Generic

The specific, fixable reasons AI output comes back bland — and what to write instead. Each mistake includes a before-and-after example you can apply immediately.

Chain-of-Thought Prompting: When It Helps and When It Wastes Money

Chain-of-thought prompting improves accuracy on reasoning tasks by asking the model to work step by step. Here is why it works, where it helps most, and when it is pure overhead.

How to Become a Prompt Engineer in 2026 (Honestly)

Whether prompt engineering is a real career, what the work actually involves, the skills that matter, and a realistic path in — including why the pure-prompting job title is fading.

Context Engineering vs Prompt Engineering: What Actually Changed

Context engineering is the term that replaced prompt engineering in serious AI work. It is not a rebrand — the job genuinely shifted from writing better instructions to controlling what the model can see. Here is the difference and why it matters.

Context window

The context window is the maximum number of tokens a model can consider at once — your prompt, any attached documents, the conversation history, and the response it generates. Exceed it and the earliest content gets dropped.

Prompt

A prompt is the input you give an AI model to produce an output. Effective prompts specify five things: the role the model should adopt, relevant context, the specific task, the output format, and any constraints.

Prompt engineering

Prompt engineering is the practice of designing model inputs that reliably produce the output you want. It combines clear instruction-writing, structured formatting, worked examples, and systematic testing.

System prompt

A system prompt is a standing instruction that shapes a model's behaviour across an entire conversation, separate from individual user messages. It typically defines persona, rules, tone and output format.

Zero-shot prompting

Zero-shot prompting means asking a model to do a task with no worked examples — just an instruction. It works well for common tasks the model has seen extensively in training.

Few-shot prompting

Few-shot prompting means including two to five worked examples of the input-output pattern you want before making your real request. It is the fastest way to lock in a consistent format, tone or edge-case behaviour.

Chain-of-thought prompting

Chain-of-thought prompting asks a model to work through its reasoning step by step before answering. It measurably improves accuracy on maths, logic, and multi-step problems by forcing intermediate steps instead of an immediate guess.

Tree-of-thought

Tree-of-thought extends chain-of-thought by exploring several reasoning branches in parallel, evaluating each, and pursuing the most promising. It helps on problems where the first plausible path is often wrong.

Self-consistency

Self-consistency runs the same reasoning prompt several times at non-zero temperature and takes the most common answer. It trades cost for accuracy on problems with a single verifiable result.

Meta-prompting

Meta-prompting is using a model to write, critique or improve prompts. You describe the task and the failure modes you want to avoid, and the model drafts the prompt.

Prompt chaining

Prompt chaining splits a complex task into a sequence of focused prompts, where each step's output feeds the next. Chains are easier to debug and more reliable than one large prompt.

Structured output

Structured output means constraining a model to return data in a machine-readable shape such as JSON, usually by specifying the exact schema in the prompt or using a provider feature that enforces it.

Function calling

Function calling lets a model request that your code run a specific function with specific arguments. You expose tool definitions; the model returns a structured call, your code executes it, and the result goes back into the conversation.

Reasoning model

A reasoning model is trained to generate extended internal deliberation before answering. It trades latency and cost for substantially better performance on maths, logic, coding and multi-step problems.

Fine-tuning

Fine-tuning continues training a pre-trained model on your own examples to specialise its behaviour. It is the right tool for teaching consistent style, format or classification behaviour — not for adding knowledge.

RAG (Retrieval-Augmented Generation)

RAG retrieves relevant passages from your own documents and inserts them into the prompt before the model answers. It grounds responses in your data, cuts hallucination, and needs no retraining.

Guardrails

Guardrails are the constraints that keep an AI system inside acceptable behaviour — input validation, output filtering, topic restrictions, action limits and human approval gates.

Prompt caching

Prompt caching stores the processed form of a repeated prompt prefix so subsequent requests reusing it are cheaper and faster. Cached input tokens typically cost a fraction of normal input tokens.

Prompt injection

Prompt injection is an attack where malicious instructions hidden in content the model processes override the developer's intended behaviour. It is the most serious unsolved security issue in LLM applications.

Model drift

Model drift is when a hosted model's behaviour changes over time as the provider updates it, causing prompts that worked reliably to degrade without any change on your side.

JSON Schema

JSON Schema is a standard vocabulary for describing the shape of JSON data — its keys, types and required fields. AI providers use it to constrain model output to a guaranteed structure.

Context engineering

Context engineering is the practice of deciding what information enters a model's context window and in what order. As windows grew, choosing what to include became more consequential than wording the instruction.

The 5-Part Prompt Framework — ChatGPT

Every strong prompt combines five parts: Role (who the AI should act as), Context (background info it needs), Task (the specific ask), Format (how you want the output structured), and Constraints (length, tone, things to avoid).

The 5-Part Prompt Framework — Claude

Every strong prompt combines five parts: Role (who the AI should act as), Context (background info it needs), Task (the specific ask), Format (how you want the output structured), and Constraints (length, tone, things to avoid).

Zero-Shot vs Few-Shot Prompting — ChatGPT

Zero-shot means asking directly with no examples.

Zero-Shot vs Few-Shot Prompting — Grok

Zero-shot means asking directly with no examples.

Chain-of-Thought Prompting — ChatGPT

Asking a model to "think step by step" before answering measurably improves accuracy on reasoning, math, and multi-step logic tasks because it forces the model to externalize intermediate steps instead of jumping to a guess..

Chain-of-Thought Prompting — Claude

Asking a model to "think step by step" before answering measurably improves accuracy on reasoning, math, and multi-step logic tasks because it forces the model to externalize intermediate steps instead of jumping to a guess..

Structured Output (JSON & Tables) — ChatGPT

When you need output your app or spreadsheet can actually use, explicitly request a schema.

Structured Output (JSON & Tables) — Hugging Face (open-source)

When you need output your app or spreadsheet can actually use, explicitly request a schema.

ChatGPT vs Claude vs Perplexity vs Grok — Perplexity

The same task often needs a slightly different prompt style per model.

ChatGPT vs Claude vs Perplexity vs Grok — Grok

The same task often needs a slightly different prompt style per model.

Role + Constraints to Kill Vagueness — ChatGPT

When answers are generic, the fix is almost always more specificity, not a longer prompt.

Role + Constraints to Kill Vagueness — Claude

When answers are generic, the fix is almost always more specificity, not a longer prompt.

Ask-Before-Answer for Ambiguous Tasks

For open-ended requests, instruct the model to ask clarifying questions first.

Self-Grading With a Rubric

Have the model score its own draft against an explicit rubric, then revise to hit the top score.

Prompt Chaining Architectures

Experts rarely use one mega-prompt.

Meta-Prompting: Prompts That Write Prompts

The fastest way to a great prompt is asking the model to write and critique one.

Evaluating Prompts Like an Engineer

A prompt is good when it performs consistently, not when one output impressed you.

Project: A 3-Step Prompt Chain for Your Real Work

Deliverable: a documented, tested 3-step prompt chain that automates one recurring task from your actual work or studies. Steps: 1.

Building a Reusable Brand Voice Prompt

Instead of re-describing your tone every time, create a single "voice block" you paste at the top of every content prompt.