> [!CAUTION]
>
> Vertex AI Feature Store (Legacy) is
> [deprecated](https://docs.cloud.google.com/vertex-ai/docs/deprecations). Beginning on
> May 17, 2026, no new features will be added and only critical
> patches will be provided. On February 17, 2027, the service will be fully sunset and APIs will
> no longer be available.
>
>
> For continued support and faster innovation, migrate to
> [Vertex AI Feature Store (V2)](https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/overview),
> our integrated platform for machine learning (ML) feature management introduced on November 17,
> 2023.

<br />

Setup includes information about setting up a project for
Vertex AI Feature Store (Legacy) and the required permissions for using
Vertex AI Feature Store (Legacy).

## Configure project

The following procedure describes how to create a new project and enable the
Vertex AI API. This API is required to use
Vertex AI Feature Store (Legacy). If you already have an existing project with
the Vertex AI API enabled, you can use that project instead of
creating a new project.

## Vertex AI Feature Store (Legacy) service agent

In addition to user permissions, Vertex AI Feature Store (Legacy) acts on your
behalf to perform operations such as accessing source data. To do so,
Vertex AI Feature Store (Legacy) uses a service agent:
`service-PROJECT_NUMBER@gcp-sa-aiplatform.iam.gserviceaccount.com`.
By default, the service agent grants Vertex AI Feature Store (Legacy) access
to source data in the same project where your featurestore is located. If the
source data is in a different project from your featurestore, you must grant the
service agent permission to access the project where the source data is
located.

For more information, see [Grant Vertex AI service agents access to other
resources](https://docs.cloud.google.com/vertex-ai/docs/general/access-control#grant_service_agents_access_to_other_resources).

## IAM permissions

Vertex AI admins have Vertex AI Feature Store (Legacy) administrator
privileges. If you require more granularity, Vertex AI Feature Store (Legacy)
provides a set of predefined IAM roles. These roles provide
different sets of permissions that are based on the following personas:

IT operations and DevOps
:   IT operations and DevOps manage Google Cloud resources and are responsible for
    creating featurestores and tuning their performance. You can use the
    `featurestoreAdmin` or `featurestoreInstanceCreator` role. The instance creator
    role lets you manage featurestores but prevents you from viewing data or
    writing data to the featurestores.

Data scientists and data engineers
:   Data scientists and data engineers create features and write data to
    featurestores. You can use the `featurestoreResourceEditor` role to
    manage entity types and features, and use the `featurestoreDataWriter` role to
    read and write feature values.

ML researchers and business analysts
:   ML researchers and business analysts search for features and export values for
    training models or making predictions; they don't need to create new features or
    write data. You can use the `featurestoreResourceViewer` role to search
    or browse for features and the `featurestoreDataViewer` role to read feature
    values.

For descriptions of each role and their associated permissions, see
[Predefined roles for
Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/general/access-control#predefined-roles).

## Quotas and limits

Vertex AI Feature Store (Legacy) enforces quotas and limits to help you manage
resources by setting your own usage limits and to protect the community of
Google Cloud users by preventing unforeseen spikes in usage. To prevent you from
hitting unplanned constraints, review Vertex AI Feature Store (Legacy) quotas
on the [Quotas and limits](https://docs.cloud.google.com/vertex-ai/quotas#featurestore) page. For example,
Vertex AI Feature Store (Legacy) sets a quota on the number of online serving
nodes and a quota on the number of online serving requests that you can make per
minute.

## What's next

- Learn about [Manage featurestores](https://docs.cloud.google.com/vertex-ai/docs/featurestore/managing-featurestores).
- Learn about [best practices](https://docs.cloud.google.com/vertex-ai/docs/featurestore/best-practices) for using Vertex AI Feature Store (Legacy).