Give your LLM the right context with RAG.
Build a retrieval pipeline you can inspect, from document preparation to evidence-backed answers.
Read the guideNeural Notes / Category
Foundations field guides.
Understand the interface before you build the integration. Start with neural API fundamentals, then follow the path from embeddings to retrieval-backed answers.
These guides focus on the agreements between components: what the caller sends, what a model-backed capability returns, and which evidence the application needs before it accepts an answer. Read the neural API introduction first when terminology is unfamiliar. Then use the retrieval guide to follow one specific application from source documents through indexing and answer generation.
A useful foundations exercise is to describe one task without mentioning a provider. Write its input, expected result, missing-information behavior, and review path. This exposes the product decisions that a convenient SDK cannot make for you. Keep that description beside you as you read the technical examples.
The articles deliberately separate model capability from service operation and from the wider product experience. That distinction matters when comparing approaches: a plausible answer, a valid response object, and a completed user task are different outcomes. Return to the glossary for a term, or continue to the LLM API track for a more focused reading path.
Guides in this collection.
What is a neural API? Start with the interface.
Models, interfaces, and the decisions that turn an interesting experiment into a reviewable application.
Read the guideTake one idea into your own project.
Write a small example of the task, a result you would accept, and a result you would reject. Keep that example beside the guide as you read. The next track connects these concepts to a practical design boundary without requiring a provider account.
Continue the reading path