---
title: "Rebuild AI search data"
method: POST
path: "/api/v1/ai/organization/projections/rebuild"
tags: ["AI Search Data"]
---

# Rebuild AI search data

`POST /api/v1/ai/organization/projections/rebuild`

## Request

- Base URL: `https://www.gooddata.ai`
- URL: `https://www.gooddata.ai/api/v1/ai/organization/projections/rebuild`
- Auth: none declared

Rebuilds the organization's AI search data from its stored metadata, memory items and knowledge document files, for example after the vector store was lost. The rebuild runs in the background; the response reports what was queued. Rebuilding knowledge documents counts toward AI usage. If this deployment cannot rebuild knowledge documents, the rebuild skips them. It still rebuilds all other AI search data.

## Response `202`

Successful Response

- ProjectionRebuildResponse: What an accepted rebuild enqueued; the workers restore it in the background.
  - `knowledgeDocuments` integer, required: Knowledge documents queued for rebuilding from their stored files.
  - `memoryItems` integer, required: Memory items queued for re-indexing.
  - `nodes` integer, required: Analytical objects queued for re-indexing.
  - `skippedKnowledgeDocuments` integer, required: Knowledge documents the rebuild skipped. A document is skipped when it has no stored file or its file is in storage the rebuild cannot read. All documents are skipped when this deployment cannot rebuild them.

## Other responses

- `404`: AI search data recovery is not enabled for the organization.
- `409`: The previous rebuild is still restoring knowledge documents.
- `429`: A rebuild was accepted recently. The Retry-After header says when to retry.

---

[API](https://skmtc.dev/gooddata/apis/gooddata-cn-oapi-all-api.md) - [All operations](https://skmtc.dev/gooddata/apis/gooddata-cn-oapi-all-api/llms.txt) - [OpenAPI document](https://skmtc.dev/gooddata/apis/gooddata-cn-oapi-all-api/revisions/c728ef303bfb?raw)