AI Unboxed
How AI really works, explained for testers.
30 “How” questions about LLMs, RAG, agents, evaluation and LLMOps. Each one starts with an everyday picture you won’t forget, then the same idea in QA words, a real example and the checks a tester runs.
0 of 30 read
AI Fundamentals
How the model reads, thinks, remembers and charges you. 6 questions.
- 01How does an LLM generate an answer, token by token?It doesn't write sentences. It predicts the next piece, again and again.Beginner✓
- 02How does tokenization work inside ChatGPT or Claude?Models don't read letters or words. They read numbered chunks.Beginner✓
- 03How does AI understand the meaning of a sentence using embeddings?Meaning becomes a location on a giant map.Beginner✓
- 04How does an LLM remember earlier messages in a conversation?Spoiler: it doesn't. The app re-sends the whole chat every time.Beginner✓
- 05How does the context window work, and what happens when it becomes full?The model's desk has a fixed size. Something must fall off.Beginner✓
- 06How can we reduce token consumption and API costs?Every token is a coin. Stop throwing coins away.Beginner✓
RAG & Knowledge Retrieval
How AI finds the right information before it answers. 8 questions.
- 07How does RAG work internally, step by step?Look it up first, then answer — an open-book exam for AI.Intermediate✓
- 08How does AI convert documents into embeddings?From messy PDF to clean numbers — the ingestion pipeline.Intermediate✓
- 09How does a vector database find similar information?Finding nearest neighbours among millions — without checking them all.Intermediate✓
- 10How does chunk size affect RAG accuracy?Too small loses meaning. Too big drowns it.Intermediate✓
- 11How does hybrid search combine keyword and vector search?Exact words + similar meaning = fewer misses.Intermediate✓
- 12How does reranking improve retrieved results?First a quick shortlist, then a careful judge.Intermediate✓
- 13How does AI process PDFs containing text, images, and tables?A PDF is a drawing of a page, not a neat document.Intermediate✓
- 14How can we improve the accuracy of a RAG chatbot?Fix the stage that's actually failing, not the one that's easiest to change.Intermediate✓
AI Agents & Tool Calling
How AI stops just talking and starts doing. 6 questions.
- 15How does an AI Agent decide which tool to call?The model reads the tool menu and picks the best-matching dish.Intermediate✓
- 16How does a tool call actually happen between an LLM and an external API?The model never touches the API. It only writes the order slip.Intermediate✓
- 17How does MCP connect AI to Jira, GitHub, Slack, and other tools?One universal plug instead of a different charger for every device.Intermediate✓
- 18How does an AI Agent execute multiple tools in sequence?Think, act, observe — repeat until the job is done.Intermediate✓
- 19How do multiple AI Agents communicate and collaborate?A team of specialists with a manager — and a shared language.Intermediate✓
- 20How does an AI Agent handle tool failures, timeouts, and retries?Real APIs fail. Good agents fail gracefully.Intermediate✓
AI Evaluation & Security
How AI goes wrong, and how testers catch it. 6 questions.
- 21How does an LLM hallucination happen?The model is built to sound right, not to know it's right.Advanced✓
- 22How can we detect and reduce hallucinations?Detection catches the lie; reduction makes it rarer.Advanced✓
- 23How does LLM-as-a-Judge evaluate another AI's response?An AI teacher grading an AI student — with a marking scheme.Advanced✓
- 24How do RAGAS and DeepEval measure AI response quality?Under the hood: claims, questions, and judge prompts.Advanced✓
- 25How does prompt injection trick AI Agents?Hidden instructions hitch a ride inside normal-looking data.Advanced✓
- 26How do guardrails protect AI applications?Safety checkpoints before, during and after the model.Advanced✓
Advanced AI Engineering
How AI is run cheaply, safely and continuously in production. 4 questions.
- 27How does prompt caching reduce AI costs?Don't pay full price to re-read the same pages every time.Lead / Architect✓
- 28How does fine-tuning differ from RAG internally?Change the brain, or hand it a book.Lead / Architect✓
- 29How do we monitor AI Agents in production?Tests prove it worked yesterday. Monitoring tells you about today.Lead / Architect✓
- 30How do we deploy and evaluate an AI application in a CI/CD pipeline?Every prompt, model or data change goes through a quality gate.Lead / Architect✓