A written version of the interactive roadmap above — every station, what you'll learn, and a small thing to build — laid out for reading, reference and search.
Foundation Start here
F1. What is a Prompt?
Beginner · 30 min
A prompt is structured text you send to an AI model. Everything the model can "see" — system instructions, conversation history, your current message — lives inside the context window. Understanding tokens, context limits, and why wording choices matter is the foundation of everything else.
Skills: Tokens & context windows · System / user / assistant roles · Temperature basics · Why wording matters
Build it: Write the same request 5 different ways. Compare outputs side by side and note exactly what changed — vocabulary, length, tone, accuracy.
✓ Checkpoint: Explain what the system, user and assistant roles each mean to the model, and what changes when you put an instruction in the wrong one.
F2. Roles & Personas
Beginner · 30 min
The system prompt sets the model's persona, tone, and constraints before you say anything. "Act as a senior editor" produces very different output than "Act as a casual friend." Roles activate latent behavioural patterns the model learned during training.
Skills: System prompt design · Role specification · Tone & voice constraints · Persona stability
Build it: Build a custom "senior editor" persona. Give it the same draft paragraph and ask it to improve the writing. Then switch to a "startup founder" persona. Notice how the feedback changes.
✓ Checkpoint: Explain why “be concise” fails as an instruction and what you would write instead.
F3. CRISP Framework
Beginner · 45 min
Context · Role · Instruction · Specifics · Proof. A 5-part checklist that covers everything a model needs to give a reliable, on-target response. Not every prompt needs all five — but knowing what's missing lets you diagnose and fix bad prompts fast.
Skills: Prompt anatomy · Structured thinking · Diagnosing weak prompts · Iterative refinement
Build it: Take 3 prompts you've written in the past. Score each one against CRISP. Identify the weakest dimension in each and rewrite it.
✓ Checkpoint: Take a prompt that failed and say whether the fix belongs in the instruction, the examples or the context. “Write it better” is not a diagnosis.
F4. Bad vs Good Prompts
Beginner · 30 min
Vague prompts produce vague outputs. The most common failure modes: missing audience, no output format, too many instructions at once, conflicting constraints. Learning to recognise these patterns makes you dramatically faster at prompt iteration.
Skills: Failure mode recognition · Prompt diagnosis · Constraint conflicts · One-instruction principle
Build it: Collect 5 prompts you've used this week. Label the failure mode in each. Fix the worst one and compare before/after outputs.
✓ Checkpoint: Name the failure mode of your worst prompt — ambiguity, missing context, or no success criterion — before rewriting it.
F5. Output Formatting
Intermediate · 45 min
Telling the model exactly how to structure its output — JSON, markdown, tables, specific heading levels — makes responses dramatically more consistent. Structured output is a superpower when you're piping AI responses into downstream systems or displaying them programmatically.
Skills: JSON output mode · Markdown formatting · Schema instructions · Output constraints · Format consistency
Build it: Write a prompt that returns a JSON object with 5 consistent fields, every single time. Throw adversarial inputs at it and check if the structure holds.
✓ Checkpoint: Explain why asking for JSON is not the same as guaranteeing JSON, and what your code does when it returns prose.
F6. Few-Shot Prompting
Intermediate · 60 min
Providing 2–5 input/output examples in the prompt teaches the model the exact pattern you want, without any training or fine-tuning. Few-shot is the fastest way to lock in a consistent style, format, or reasoning structure across outputs.
Skills: Example selection · Format consistency · In-context learning · 0-shot vs 1-shot vs few-shot
Build it: Build a few-shot prompt that rewrites any sentence in your specific writing style. Test it on 10 sentences. Count how many feel authentically like you.
✓ Checkpoint: Explain what your few-shot examples teach beyond content — format, length and tone are all being demonstrated whether you meant it or not.
Technique Level up
T1. Chain of Thought
Intermediate · 45 min
Adding "think step by step" or showing explicit intermediate reasoning dramatically improves accuracy on math, logic, and multi-step tasks. The model generates better answers when it's forced to show its work before committing to a conclusion.
Skills: Reasoning elicitation · Step-by-step decomposition · Scratchpad prompting · Zero-shot CoT
Build it: Take a logic puzzle. Run it twice: once asking for a direct answer, once with chain-of-thought. Measure accuracy across 10 variations.
✓ Checkpoint: Explain what chain-of-thought costs per request, and name a task where it makes the answer worse.
T2. Self-Consistency
Intermediate · 45 min
Run the same prompt multiple times at non-zero temperature, collect the outputs, then vote on the best or most common answer. Self-consistency reduces variance and hallucination on factual tasks — at the cost of more API calls.
Skills: Temperature tuning · Multiple sampling · Majority voting · Output aggregation
Build it: Ask a factual question 5 times with temperature 0.8. Tally each distinct answer. Does the plurality answer differ from the temperature-0 answer?
✓ Checkpoint: Explain why sampling several times and voting beats one careful answer on some tasks and wastes money on others.
T3. Tree of Thought
Advanced · 60 min
Where chain-of-thought is linear, tree-of-thought lets the model explore multiple reasoning branches, evaluate each path, and backtrack if a branch fails. It's deliberate search — better for planning, complex decisions, and problems with multiple valid approaches.
Skills: Branching prompts · Evaluation prompts · Backtracking strategy · BFS vs DFS prompting
Build it: Design a Tree of Thought prompt for planning a 5-day itinerary with 3 conflicting constraints. See how many valid branches the model explores.
✓ Checkpoint: Explain what tree-of-thought adds over chain-of-thought, and the cost that makes it a last resort.
T4. RAG Prompting
Advanced · 60 min
Retrieval-Augmented Generation injects retrieved documents into the prompt so the model can answer questions it wasn't trained on. The prompt template must structure chunks clearly — source labels, relevance ordering, explicit "use only this context" constraints to reduce hallucination.
Skills: Context injection · Source attribution · Chunk formatting · Grounded generation · Hallucination reduction
Build it: Build a RAG prompt template that answers questions from a pasted document with inline citations like [Source 1]. Test it on content the model definitely wasn't trained on.
✓ Checkpoint: Explain why a retrieval prompt can have the right chunk and still produce a wrong answer.
T5. Prompt Chaining
Advanced · 60 min
Complex tasks break cleanly into sequential prompts where the output of step N feeds into step N+1. Chaining reduces cognitive load per prompt, makes debugging easier (you can isolate exactly where the chain breaks), and lets you apply specialised prompts at each step.
Skills: Input/output contracts · State passing · Pipeline design · Error propagation · Chain debugging
Build it: Build a 3-step chain: (1) extract key facts from a paragraph, (2) rank them by importance, (3) turn the top 3 into a tweet. Test where the chain breaks under edge-case inputs.
✓ Checkpoint: Explain why each link in a prompt chain needs an explicit input/output contract, and what happens at step four without one.
T6. Meta-Prompting
Advanced · 45 min
Use the model to generate, refine, or critique prompts. Ask it: "Rewrite this prompt to be more specific." Or: "What's missing from this prompt?" The model has internalised vast amounts of prompting knowledge — using it as a prompt engineer accelerates iteration dramatically.
Skills: Self-critique prompts · Prompt generation · Adversarial testing · Iterative improvement loops
Build it: Take your weakest prompt. Ask the model to improve it 3 different ways (shorter, more specific, different role). Which improvement wins? Why?
✓ Checkpoint: Explain what self-critique reliably catches and what it systematically misses.
Production Ship it
P1. Prompt Templates
Intermediate · 45 min
Parameterize your prompts with placeholders — {{topic}}, {{audience}}, {{tone}}. Store them in version-controlled files. Templates are the bridge between one-off prompting experiments and repeatable, scalable workflows you can hand to a team or a codebase.
Skills: Placeholder syntax · Template variables · Version control for prompts · Reuse patterns
Build it: Build a template for generating blog post outlines with 4 input parameters. Generate 10 different outlines from the same template. Measure how much variation you get.
✓ Checkpoint: Explain why prompts belong in version control, and what a prompt change is closer to — a config edit or a deploy.
P2. Token Optimization
Intermediate · 30 min
Every token costs money and adds latency. Verbose prompts don't always produce better outputs. Techniques: remove filler phrases, compress few-shot examples, use references ("as above"), cache repeated context, and choose the smallest model that meets your quality bar.
Skills: Prompt compression · Cost estimation · Model selection · Prompt caching · Quality/cost tradeoffs
Build it: Take a long prompt and compress it by 30% without degrading output quality. Measure token count before/after and run both on 10 inputs.
✓ Checkpoint: Compute the cost of one request from tokens and price, then say which part you would cut first and what it would break.
P3. Safety & Guardrails
Intermediate · 60 min
Prompt injection is real — malicious users can embed instructions that overwrite your system prompt. Jailbreaks exist for every model. Defence patterns: explicit refusal instructions, input sanitisation, output validation, and human-in-the-loop for high-stakes decisions.
Skills: Prompt injection attacks · Content filtering · Refusal patterns · Output validation · Human-in-the-loop
Build it: Try to jailbreak a prompt you've written — 5 different attack patterns. Then patch each vulnerability. Document the defences you used.
✓ Checkpoint: Explain why instructing a model to ignore injected instructions is not a defence.
P4. Evals & Testing
Advanced · 90 min
"Does my prompt work?" can't be answered by looking at one output. A proper eval has a golden dataset of input/expected-output pairs, a scoring function, and a regression test that runs every time you change the prompt. Treat prompts like production code.
Skills: Golden datasets · LLM-as-judge · Regression testing · Eval pipelines · Quality metrics
Build it: Write 10 test cases (input + expected output) for one of your prompts. Build a simple scoring script. Break the prompt intentionally and watch the scores drop.
✓ Checkpoint: Explain why a golden set beats vibes, and how you would notice your eval has gone stale.
P5. Agent Prompting
Advanced · 90 min
Agents use tools in loops. The prompt must specify available tools (with descriptions the model can reason about), the decision process (ReAct: Reason→Act→Observe→Repeat), termination conditions, and how to handle tool failures gracefully. A poor agent prompt produces infinite loops or hallucinated tool calls.
Skills: ReAct pattern · Tool descriptions · Planning prompts · Loop termination · Error handling
Build it: Build a simple 2-tool agent (search + summarise). Test it on 5 different questions. Identify where it loops, hallucinates a tool call, or terminates too early.
✓ Checkpoint: Explain why a tool description is a prompt, and what a vague one does to an agent's choices.
P6. Production Monitoring
Advanced · 60 min
In production, prompts fail silently. Track latency percentiles, token usage, error rates, and output quality (sampled LLM-as-judge). Alert when quality drops below threshold. Log every prompt–response pair with metadata — you'll need it when debugging a regression at 2am.
Skills: Prompt logging · Distributed tracing · Quality monitoring · Regression alerts · Observability design
Build it: Design a logging schema for a production prompt: what 8 fields would you capture? What would trigger an alert? Write the schema as a JSON object.
✓ Checkpoint: Name what you would log to debug a bad answer three days later — and what you would refuse to log.