Beginner
AI Prompting
#LLM
#RAG
#embeddings
#fine-tuning
AI Vocabulary
4 exercises — master the essential terms every developer needs to discuss AI: LLM, RAG, tokens, context window, fine-tuning, and more.
0 / 24 completed
1 / 24
What is a Large Language Model (LLM)?
An LLM (Large Language Model) is a type of deep learning model trained on very large text datasets. It learns statistical patterns in language and can:
• Generate text — write code, emails, articles, summaries
• Answer questions — based on patterns learned during training
• Translate languages
• Reason over context — analyse documents, compare options, explain concepts
Examples: GPT-4 (OpenAI), Claude (Anthropic), Gemini (Google), Llama (Meta), Mistral.
Key vocabulary:
• Parameters — the numbers inside the model that were trained. GPT-4 has an estimated ~1 trillion parameters.
• Pre-training — the initial training phase on massive text data
• Fine-tuning — additional training on a smaller dataset to specialise behaviour
• Inference — using the trained model to generate a response (the "running" phase, as opposed to training)
• Generate text — write code, emails, articles, summaries
• Answer questions — based on patterns learned during training
• Translate languages
• Reason over context — analyse documents, compare options, explain concepts
Examples: GPT-4 (OpenAI), Claude (Anthropic), Gemini (Google), Llama (Meta), Mistral.
Key vocabulary:
• Parameters — the numbers inside the model that were trained. GPT-4 has an estimated ~1 trillion parameters.
• Pre-training — the initial training phase on massive text data
• Fine-tuning — additional training on a smaller dataset to specialise behaviour
• Inference — using the trained model to generate a response (the "running" phase, as opposed to training)
2 / 24
Your team is building a feature where an LLM answers questions about your company's internal documentation. Which technique best describes this approach?
RAG (Retrieval-Augmented Generation) is the technique of combining an LLM with a search/retrieval system so the model can answer questions based on external or private documents.
How RAG works:
1. User asks a question
2. A retrieval system searches a document store (vector database, Elasticsearch, etc.) for relevant chunks
3. The relevant chunks are injected into the LLM's prompt as context
4. The LLM generates an answer grounded in the retrieved documents
Why RAG over fine-tuning for this use case?
• Fine-tuning embeds knowledge into model weights — expensive, slow to update, requires training data
• RAG retrieves fresh documents at query time — cheap to update (just update the document store), no retraining needed
Key RAG vocabulary:
• Embeddings — vector representations of text used for similarity search
• Vector database — a database optimised to store and search embeddings (e.g. Pinecone, Weaviate, pgvector)
• Chunk — a piece of the document that is retrieved and injected
• Context window — the maximum amount of text the LLM can process at once (determines how many chunks fit)
How RAG works:
1. User asks a question
2. A retrieval system searches a document store (vector database, Elasticsearch, etc.) for relevant chunks
3. The relevant chunks are injected into the LLM's prompt as context
4. The LLM generates an answer grounded in the retrieved documents
Why RAG over fine-tuning for this use case?
• Fine-tuning embeds knowledge into model weights — expensive, slow to update, requires training data
• RAG retrieves fresh documents at query time — cheap to update (just update the document store), no retraining needed
Key RAG vocabulary:
• Embeddings — vector representations of text used for similarity search
• Vector database — a database optimised to store and search embeddings (e.g. Pinecone, Weaviate, pgvector)
• Chunk — a piece of the document that is retrieved and injected
• Context window — the maximum amount of text the LLM can process at once (determines how many chunks fit)
3 / 24
An LLM has a context window of 128,000 tokens. Roughly how many words is that?
Option C is correct. The common rule of thumb is: 1 token ≈ 0.75 words (in English). So 128,000 tokens ≈ 96,000 words.
For reference: the average novel is ~80,000–100,000 words. A 128K context window can hold roughly one entire novel.
Why tokens, not words?
LLMs don't process words — they process tokens, which are sub-word units. The word "tokenisation" might be split into ["token", "isation"] = 2 tokens. Code is typically more token-dense than prose (braces, semicolons, and operators each cost tokens).
Practical implications:
• Long documents — if a document exceeds the context window, you must truncate, chunk (RAG), or summarise
• Cost — most LLM APIs charge per token (input + output tokens combined)
• Temperature (not related to context, but often asked together) — a setting (0 to 1) controlling output randomness. Temperature 0 = deterministic/consistent; Temperature 1 = creative/varied
For reference: the average novel is ~80,000–100,000 words. A 128K context window can hold roughly one entire novel.
Why tokens, not words?
LLMs don't process words — they process tokens, which are sub-word units. The word "tokenisation" might be split into ["token", "isation"] = 2 tokens. Code is typically more token-dense than prose (braces, semicolons, and operators each cost tokens).
Practical implications:
• Long documents — if a document exceeds the context window, you must truncate, chunk (RAG), or summarise
• Cost — most LLM APIs charge per token (input + output tokens combined)
• Temperature (not related to context, but often asked together) — a setting (0 to 1) controlling output randomness. Temperature 0 = deterministic/consistent; Temperature 1 = creative/varied
4 / 24
What is the difference between fine-tuning and prompting an LLM?
Option B is the correct and important distinction:
Prompting (zero-shot, few-shot, chain-of-thought)
• Instructions given in the conversation at query time
• No model weights are modified
• Available to everyone, no ML expertise needed
• Effect lasts only for that conversation
• Cost: only the token cost of the request
Fine-tuning
• Continues the training process with new labelled data
• Updates the model's weights — the knowledge or style is baked in permanently
• Requires a training dataset, ML infrastructure, and GPU time
• Permanently changes how the model behaves (for that version)
• Cost: training compute (significant) + serving the fine-tuned model
When to use each:
• Use prompting for: context injection, format control, task steering
• Use fine-tuning for: consistent style/tone that prompting can't achieve, domain-specific knowledge that must be in weights, faster inference by reducing prompt length
Most production AI features start with prompting and only fine-tune when the use case demands it.
Prompting (zero-shot, few-shot, chain-of-thought)
• Instructions given in the conversation at query time
• No model weights are modified
• Available to everyone, no ML expertise needed
• Effect lasts only for that conversation
• Cost: only the token cost of the request
Fine-tuning
• Continues the training process with new labelled data
• Updates the model's weights — the knowledge or style is baked in permanently
• Requires a training dataset, ML infrastructure, and GPU time
• Permanently changes how the model behaves (for that version)
• Cost: training compute (significant) + serving the fine-tuned model
When to use each:
• Use prompting for: context injection, format control, task steering
• Use fine-tuning for: consistent style/tone that prompting can't achieve, domain-specific knowledge that must be in weights, faster inference by reducing prompt length
Most production AI features start with prompting and only fine-tune when the use case demands it.
5 / 24
Sarah: 'The prompt engineering for this new chatbot is proving more complex than anticipated. We're struggling to get consistent answers from the LLM regarding our API authentication flows. It seems like it's frequently hallucinating details or misinterpreting the documentation snippets we provide.'
Which of the following phrases best reflects Sarah's concern, demonstrating a key challenge in AI prompting?Sarah correctly identifies the core problem: the LLM lacks sufficient grounding in the provided documentation. 'Hallucination' refers to the model generating plausible but entirely false information – a common issue when prompts aren't carefully crafted and don't provide enough context or constraints. Options A and B focus on technical parameters (temperature) or hardware limitations, while option C directly addresses the root cause of the problem and uses accurate terminology. Choosing the incorrect LLM is a reactive solution rather than addressing the prompt design itself.6 / 24
Reviewer: 'The response is technically correct, but the LLM's explanation lacks sufficient detail for a junior developer. It doesn't address potential edge cases or offer guidance on error handling. Perhaps we should experiment with a more verbose prompt to encourage it to elaborate.' During a code review of an API integration built using an LLM, your team lead comments on the response quality. Which phrase most accurately captures the core issue being raised?
This scenario highlights a critical aspect of AI prompting: ensuring the LLM provides *sufficient* detail for its intended audience. The reviewer isn't simply criticizing that the response is 'wrong,' but rather that it's lacking the necessary depth and context for developers to effectively use it – especially junior ones. The key misconception here is assuming a technically correct answer automatically equates to a useful one; prompt engineering needs to guide the LLM toward comprehensive responses, anticipating potential user needs and complexities.
7 / 24
Reviewer: 'The LLM's response is factually accurate according to the documentation, but it consistently fails to surface potential security vulnerabilities when prompted about user authentication. It seems to prioritize brevity over comprehensive risk assessment – a critical oversight for our application.' During a code review of an API integration built using an LLM, your team lead provides this feedback. Which phrase best describes the primary concern being expressed?The reviewer highlights a crucial failure of the LLM: neglecting to identify security vulnerabilities. The incorrect options focus on issues like brevity or hallucination, which are related but don't capture the core problem – the LLM isn't performing its intended risk assessment function. This demonstrates a common challenge in AI prompting: models can be accurate without being *useful* when addressing critical requirements like security, requiring more carefully crafted prompts and potentially post-processing of outputs.
8 / 24
During a Slack discussion about integrating an LLM into a new customer support chatbot, Alex writes: 'I'm seeing the model consistently generate responses that are *too confident* and outright incorrect. It's like it's inventing details to fill in gaps in our knowledge base – we need better ways to control its output and ensure accuracy.' Which of the following phrases BEST captures the crux of Alex's concern regarding AI prompting?
Alex's concern centers on the LLM's tendency to fabricate information – a phenomenon called 'hallucination.' This highlights a key challenge in AI prompting: ensuring the model doesn't confidently present incorrect or misleading answers. The correct option directly addresses this issue of inaccurate generation stemming from insufficient prompt control, not general training data limitations or architectural problems.
9 / 24
Reviewer: 'The LLM's response is technically correct according to the documentation, but it's consistently generating overly optimistic estimates for server load during peak hours. It's suggesting we scale up by 50% without considering historical data or current traffic patterns – a classic example of *hallucination* in prompt engineering. This could lead to significant over-provisioning and wasted resources.' During a code review of an API integration built using an LLM, your team lead identifies this issue. Which phrase best reflects the core problem being highlighted regarding the LLM's output?This scenario illustrates 'hallucination,' a common challenge in LLMs where they generate plausible-sounding but factually incorrect information. The reviewer isn't just criticizing the response for being wrong; they're pointing out that the LLM is confidently presenting misleading data as truth – a direct consequence of the model extrapolating without sufficient grounding. Options A, C and D represent potential issues with the prompt or broader system design, but the core problem here is the *hallucination* itself.
10 / 24
Sarah: 'The prompt engineering for this new chatbot is proving more complex than anticipated. We're struggling to get consistent answers from the LLM regarding our API authentication flows. It seems like it's frequently hallucinating details or misinterpreting the documentation snippets we provide.'
Which of the following phrases best reflects Sarah's concern, demonstrating a key challenge in AI prompting?Sarah correctly identifies the core problem: the LLM lacks sufficient grounding in the provided documentation. 'Hallucination' refers to the model generating plausible but entirely false information – a common issue when prompts aren't carefully crafted and don't provide enough context or constraints. Options A and B focus on technical parameters (temperature) or hardware limitations, while option C directly addresses the root cause of the problem and uses accurate terminology. Choosing the incorrect LLM is a reactive solution rather than addressing the prompt design itself.11 / 24
Reviewer: 'The response is technically correct, but the LLM's explanation lacks sufficient detail for a junior developer. It doesn't address potential edge cases or offer guidance on error handling. Perhaps we should experiment with a more verbose prompt to encourage it to elaborate.' During a code review of an API integration built using an LLM, your team lead comments on the response quality. Which phrase most accurately captures the core issue being raised?
This scenario highlights a critical aspect of AI prompting: ensuring the LLM provides *sufficient* detail for its intended audience. The reviewer isn't simply criticizing that the response is 'wrong,' but rather that it's lacking the necessary depth and context for developers to effectively use it – especially junior ones. The key misconception here is assuming a technically correct answer automatically equates to a useful one; prompt engineering needs to guide the LLM toward comprehensive responses, anticipating potential user needs and complexities.
12 / 24
Reviewer: 'The LLM's response is factually accurate according to the documentation, but it consistently fails to surface potential security vulnerabilities when prompted about user authentication. It seems to prioritize brevity over comprehensive risk assessment – a critical oversight for our application.' During a code review of an API integration built using an LLM, your team lead provides this feedback. Which phrase best describes the primary concern being expressed?The reviewer highlights a crucial failure of the LLM: neglecting to identify security vulnerabilities. The incorrect options focus on issues like brevity or hallucination, which are related but don't capture the core problem – the LLM isn't performing its intended risk assessment function. This demonstrates a common challenge in AI prompting: models can be accurate without being *useful* when addressing critical requirements like security, requiring more carefully crafted prompts and potentially post-processing of outputs.
13 / 24
During a Slack discussion about integrating an LLM into a new customer support chatbot, Alex writes: 'I'm seeing the model consistently generate responses that are *too confident* and outright incorrect. It's like it's inventing details to fill in gaps in our knowledge base – we need better ways to control its output and ensure accuracy.' Which of the following phrases BEST captures the crux of Alex's concern regarding AI prompting?
Alex's concern centers on the LLM's tendency to fabricate information – a phenomenon called 'hallucination.' This highlights a key challenge in AI prompting: ensuring the model doesn't confidently present incorrect or misleading answers. The correct option directly addresses this issue of inaccurate generation stemming from insufficient prompt control, not general training data limitations or architectural problems.
14 / 24
Reviewer: 'The LLM's response is technically correct according to the documentation, but it's consistently generating overly optimistic estimates for server load during peak hours. It's suggesting we scale up by 50% without considering historical data or current traffic patterns – a classic example of *hallucination* in prompt engineering. This could lead to significant over-provisioning and wasted resources.' During a code review of an API integration built using an LLM, your team lead identifies this issue. Which phrase best reflects the core problem being highlighted regarding the LLM's output?This scenario illustrates 'hallucination,' a common challenge in LLMs where they generate plausible-sounding but factually incorrect information. The reviewer isn't just criticizing the response for being wrong; they're pointing out that the LLM is confidently presenting misleading data as truth – a direct consequence of the model extrapolating without sufficient grounding. Options A, C and D represent potential issues with the prompt or broader system design, but the core problem here is the *hallucination* itself.
15 / 24
Sarah: 'The prompt engineering for this new chatbot is proving more complex than anticipated. We're struggling to get consistent answers from the LLM regarding our API authentication flows. It seems like it's frequently hallucinating details or misinterpreting the documentation snippets we provide.'
Which of the following phrases best reflects Sarah's concern, demonstrating a key challenge in AI prompting?Sarah correctly identifies the core problem: the LLM lacks sufficient grounding in the provided documentation. 'Hallucination' refers to the model generating plausible but entirely false information – a common issue when prompts aren't carefully crafted and don't provide enough context or constraints. Options A and B focus on technical parameters (temperature) or hardware limitations, while option C directly addresses the root cause of the problem and uses accurate terminology. Choosing the incorrect LLM is a reactive solution rather than addressing the prompt design itself.16 / 24
Reviewer: 'The response is technically correct, but the LLM's explanation lacks sufficient detail for a junior developer. It doesn't address potential edge cases or offer guidance on error handling. Perhaps we should experiment with a more verbose prompt to encourage it to elaborate.' During a code review of an API integration built using an LLM, your team lead comments on the response quality. Which phrase most accurately captures the core issue being raised?
This scenario highlights a critical aspect of AI prompting: ensuring the LLM provides *sufficient* detail for its intended audience. The reviewer isn't simply criticizing that the response is 'wrong,' but rather that it's lacking the necessary depth and context for developers to effectively use it – especially junior ones. The key misconception here is assuming a technically correct answer automatically equates to a useful one; prompt engineering needs to guide the LLM toward comprehensive responses, anticipating potential user needs and complexities.
17 / 24
Reviewer: 'The LLM's response is factually accurate according to the documentation, but it consistently fails to surface potential security vulnerabilities when prompted about user authentication. It seems to prioritize brevity over comprehensive risk assessment – a critical oversight for our application.' During a code review of an API integration built using an LLM, your team lead provides this feedback. Which phrase best describes the primary concern being expressed?The reviewer highlights a crucial failure of the LLM: neglecting to identify security vulnerabilities. The incorrect options focus on issues like brevity or hallucination, which are related but don't capture the core problem – the LLM isn't performing its intended risk assessment function. This demonstrates a common challenge in AI prompting: models can be accurate without being *useful* when addressing critical requirements like security, requiring more carefully crafted prompts and potentially post-processing of outputs.
18 / 24
During a Slack discussion about integrating an LLM into a new customer support chatbot, Alex writes: 'I'm seeing the model consistently generate responses that are *too confident* and outright incorrect. It's like it's inventing details to fill in gaps in our knowledge base – we need better ways to control its output and ensure accuracy.' Which of the following phrases BEST captures the crux of Alex's concern regarding AI prompting?
Alex's concern centers on the LLM's tendency to fabricate information – a phenomenon called 'hallucination.' This highlights a key challenge in AI prompting: ensuring the model doesn't confidently present incorrect or misleading answers. The correct option directly addresses this issue of inaccurate generation stemming from insufficient prompt control, not general training data limitations or architectural problems.
19 / 24
Reviewer: 'The LLM's response is technically correct according to the documentation, but it's consistently generating overly optimistic estimates for server load during peak hours. It's suggesting we scale up by 50% without considering historical data or current traffic patterns – a classic example of *hallucination* in prompt engineering. This could lead to significant over-provisioning and wasted resources.' During a code review of an API integration built using an LLM, your team lead identifies this issue. Which phrase best reflects the core problem being highlighted regarding the LLM's output?This scenario illustrates 'hallucination,' a common challenge in LLMs where they generate plausible-sounding but factually incorrect information. The reviewer isn't just criticizing the response for being wrong; they're pointing out that the LLM is confidently presenting misleading data as truth – a direct consequence of the model extrapolating without sufficient grounding. Options A, C and D represent potential issues with the prompt or broader system design, but the core problem here is the *hallucination* itself.
20 / 24
Sarah: 'The prompt engineering for this new chatbot is proving more complex than anticipated. We're struggling to get consistent answers from the LLM regarding our API authentication flows. It seems like it's frequently hallucinating details or misinterpreting the documentation snippets we provide.'
Which of the following phrases best reflects Sarah's concern, demonstrating a key challenge in AI prompting?Sarah correctly identifies the core problem: the LLM lacks sufficient grounding in the provided documentation. 'Hallucination' refers to the model generating plausible but entirely false information – a common issue when prompts aren't carefully crafted and don't provide enough context or constraints. Options A and B focus on technical parameters (temperature) or hardware limitations, while option C directly addresses the root cause of the problem and uses accurate terminology. Choosing the incorrect LLM is a reactive solution rather than addressing the prompt design itself.21 / 24
Reviewer: 'The response is technically correct, but the LLM's explanation lacks sufficient detail for a junior developer. It doesn't address potential edge cases or offer guidance on error handling. Perhaps we should experiment with a more verbose prompt to encourage it to elaborate.' During a code review of an API integration built using an LLM, your team lead comments on the response quality. Which phrase most accurately captures the core issue being raised?
This scenario highlights a critical aspect of AI prompting: ensuring the LLM provides *sufficient* detail for its intended audience. The reviewer isn't simply criticizing that the response is 'wrong,' but rather that it's lacking the necessary depth and context for developers to effectively use it – especially junior ones. The key misconception here is assuming a technically correct answer automatically equates to a useful one; prompt engineering needs to guide the LLM toward comprehensive responses, anticipating potential user needs and complexities.
22 / 24
Reviewer: 'The LLM's response is factually accurate according to the documentation, but it consistently fails to surface potential security vulnerabilities when prompted about user authentication. It seems to prioritize brevity over comprehensive risk assessment – a critical oversight for our application.' During a code review of an API integration built using an LLM, your team lead provides this feedback. Which phrase best describes the primary concern being expressed?The reviewer highlights a crucial failure of the LLM: neglecting to identify security vulnerabilities. The incorrect options focus on issues like brevity or hallucination, which are related but don't capture the core problem – the LLM isn't performing its intended risk assessment function. This demonstrates a common challenge in AI prompting: models can be accurate without being *useful* when addressing critical requirements like security, requiring more carefully crafted prompts and potentially post-processing of outputs.
23 / 24
During a Slack discussion about integrating an LLM into a new customer support chatbot, Alex writes: 'I'm seeing the model consistently generate responses that are *too confident* and outright incorrect. It's like it's inventing details to fill in gaps in our knowledge base – we need better ways to control its output and ensure accuracy.' Which of the following phrases BEST captures the crux of Alex's concern regarding AI prompting?
Alex's concern centers on the LLM's tendency to fabricate information – a phenomenon called 'hallucination.' This highlights a key challenge in AI prompting: ensuring the model doesn't confidently present incorrect or misleading answers. The correct option directly addresses this issue of inaccurate generation stemming from insufficient prompt control, not general training data limitations or architectural problems.
24 / 24
Reviewer: 'The LLM's response is technically correct according to the documentation, but it's consistently generating overly optimistic estimates for server load during peak hours. It's suggesting we scale up by 50% without considering historical data or current traffic patterns – a classic example of *hallucination* in prompt engineering. This could lead to significant over-provisioning and wasted resources.' During a code review of an API integration built using an LLM, your team lead identifies this issue. Which phrase best reflects the core problem being highlighted regarding the LLM's output?This scenario illustrates 'hallucination,' a common challenge in LLMs where they generate plausible-sounding but factually incorrect information. The reviewer isn't just criticizing the response for being wrong; they're pointing out that the LLM is confidently presenting misleading data as truth – a direct consequence of the model extrapolating without sufficient grounding. Options A, C and D represent potential issues with the prompt or broader system design, but the core problem here is the *hallucination* itself.
Frequently Asked Questions
What will I practice in "AI Vocabulary — AI Prompting English Exercise"?
This is an AI Prompting exercise set. It walks through 24 scenario-based multiple-choice questions built around real usage of AI Prompting terminology that IT professionals encounter on the job.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to complete with no account, sign-up, or paywall.
How many questions are in this exercise?
This set contains 24 questions. Each one shows immediate feedback and a detailed explanation after you answer, so you learn the correct usage right away rather than waiting for a final score.
Do I need prior experience to complete this exercise?
No prior experience is required. Each question includes a full explanation covering the reasoning behind the correct answer, so the exercise itself teaches the AI Prompting vocabulary as you go.
Can I retry the exercise if I get questions wrong?
Yes — use the "Try again" button on the results screen to reset your answers and go through all the questions again. There is no limit on attempts.
Is my progress saved?
Your answers and score for the current session are tracked in the browser as you go. No account or login is needed, and there is nothing to install.
What if I don't understand a term used in a question?
Read the explanation shown after you answer each question — it breaks down the correct term in plain English with a real-world example. You can also check the site Glossary for quick definitions.
How is this different from reading a blog article on the topic?
Exercises like this one are interactive drills that test and reinforce specific vocabulary through multiple-choice questions, while blog articles explain concepts in prose. Practising here after reading builds active recall, not just passive recognition.
Where can I find more AI Prompting exercises?
See the AI Prompting exercises hub for the full set of related pages, or browse all exercise categories from the main Exercises index.
Can I use this exercise to prepare for a technical interview?
Yes — AI Prompting vocabulary comes up often in technical discussions and interviews. Pair this exercise with our dedicated Interview Preparation section for role-specific practice.