System messages, few-shot examples, chain of thought and grounding, and which fix suits which prompt problem.
From Ultra Transcenders AI-103 by Tony Rough (publishing soon)
Prompt engineering steers a model’s output without changing its weights, so it is always the first and cheapest thing to try.
| Technique | What it is |
|---|---|
| System message / priming | Role, persona, audience, allowed scope and refusal behaviour, set before any user input; the place to enforce a topic boundary |
| Few-shot | Example user/assistant pairs showing format and behaviour; reinforce the system message but don’t enforce a boundary |
| Chain of thought | Ask for step-by-step reasoning |
| Affordance | Give the model tools or functions to call |
| Grounding content | Supply the facts to answer from (RAG) |
This note is one section of Ultra Transcenders AI-103: Developing AI Apps and Agents on Azure, an independent study guide that explains every topic the exam covers by technology, with comparison tables, diagrams and the common traps, plus a glossary linked to Microsoft Learn.
Publishing soon on Amazon in Kindle and paperback editions.
About the book · Free AI-103 glossary · All AI-103 study notes
Where guardrails check an agent run, what Prompt Shields catch, and when to block or annotate.
File search, Azure AI Search, Bing grounding, function, OpenAPI, MCP and code interpreter tools compared.
How to keep humans in the loop and limit what an agent's tools can do.
The indexer pipeline stages, built-in and custom skills, and knowledge store projections.
Vector fields and profiles, HNSW vs exhaustive KNN, hybrid queries with RRF, and the semantic ranker.
Prebuilt and custom analyzers, field extraction methods and Markdown output for RAG.
What Sora 2 can generate, its parameters and limits, and how the asynchronous jobs work.