---
title: "Use cases"
sidebarTitle: "Overview"
description: "Worked examples of building with LanceDB, by what you are building."
---

Each page here is one worked example, grouped by what it builds. The
documentation explains how a feature works; these show it doing something.

| | |
|:---------|:------------|
| [Search](/tutorials/search/multivector-needle-in-a-haystack) | Vector, multi-vector, hybrid and full-text retrieval. |
| [RAG](/tutorials/agents/nvidia-rag-blueprint/) | Retrieval-augmented generation and agent applications. |
| [Feature engineering](/tutorials/feature-engineering/) | Deriving columns, materialized views, and experimenting on branches. |
| [Training](/training/object-detection) | Curating data and training models directly on a table. |

Each category also has a **More examples** page collecting runnable notebooks
that live in the recipes repository rather than here.

## Recipes {#recipes}

If you are looking for ideas and hands-on code, the repository below collects
practical projects beyond the ones documented here.

<Card icon="code" title="VectorDB recipes" href="https://github.com/lancedb/vectordb-recipes">
Past code examples and tutorials on GitHub.
</Card>