The GenAI Field Guide

Ship your first LLM feature

Tokens to production in the order you will need them: prompts, structure, retrieval, tools, tests and cost.

1. How does an LLM generate text?
2. What is a token?
3. What is a context window?
4. What does temperature change?
5. What is a system prompt?
6. What is few-shot prompting?
7. What is a JSON Schema?
8. What if output fails validation?
9. What is RAG?
10. What is chunking?
11. What is hybrid search?
12. What is tool calling?
13. Does the model execute a tool itself?
14. What is an eval dataset?
15. What is a regression eval?
16. Should the frontend call a model directly?
17. What is streaming?
18. How is LLM API pricing calculated?
19. What is a trace?