haive.core.engine.retriever.providers.GoogleVertexAISearchRetrieverConfig

Google Vertex AI Search Retriever implementation for the Haive framework.

from typing import Any This module provides a configuration class for the Google Vertex AI Search retriever, which uses Google Cloud’s Vertex AI Search (formerly Enterprise Search) service. Vertex AI Search provides ML-powered search capabilities with natural language understanding and enterprise-grade security and compliance.

The GoogleVertexAISearchRetriever works by: 1. Connecting to a Vertex AI Search data store 2. Executing search queries with ML understanding 3. Returning ranked results with relevance scoring 4. Supporting various data source types and formats

This retriever is particularly useful when: - Building enterprise search on Google Cloud - Need ML-powered query understanding - Working with Google Cloud data sources - Want enterprise security and compliance - Building knowledge management systems

The implementation integrates with LangChain’s GoogleVertexAISearchRetriever while providing a consistent Haive configuration interface with secure GCP credential management.

Classes

GoogleVertexAISearchRetrieverConfig

Configuration for Google Vertex AI Search retriever in the Haive framework.

Module Contents

class haive.core.engine.retriever.providers.GoogleVertexAISearchRetrieverConfig.GoogleVertexAISearchRetrieverConfig

Bases: haive.core.common.mixins.secure_config.SecureConfigMixin, haive.core.engine.retriever.retriever.BaseRetrieverConfig

Configuration for Google Vertex AI Search retriever in the Haive framework.

This retriever uses Google Cloud Vertex AI Search to provide ML-powered enterprise search with natural language understanding.

retriever_type

The type of retriever (always GOOGLE_VERTEX_AI_SEARCH).

Type:

RetrieverType

project_id

Google Cloud project ID.

Type:

str

data_store_id

Vertex AI Search data store ID.

Type:

str

location_id

Google Cloud location ID.

Type:

str

serving_config_id

Serving configuration ID.

Type:

str

api_key

Service account key (auto-resolved from GOOGLE_APPLICATION_CREDENTIALS).

Type:

Optional[SecretStr]

max_documents

Maximum number of documents to retrieve.

Type:

int

Examples

>>> from haive.core.engine.retriever import GoogleVertexAISearchRetrieverConfig
>>>
>>> # Create the Vertex AI Search retriever config
>>> config = GoogleVertexAISearchRetrieverConfig(
...     name="vertex_search_retriever",
...     project_id="my-gcp-project",
...     data_store_id="my-data-store",
...     location_id="global",
...     serving_config_id="default_config",
...     max_documents=10
... )
>>>
>>> # Instantiate and use the retriever
>>> retriever = config.instantiate()
>>> docs = retriever.get_relevant_documents("enterprise search capabilities")
>>>
>>> # Example with regional deployment
>>> regional_config = GoogleVertexAISearchRetrieverConfig(
...     name="regional_vertex_search",
...     project_id="my-gcp-project",
...     data_store_id="my-data-store",
...     location_id="us-central1",
...     serving_config_id="custom_config"
... )

get_input_fields()

Return input field definitions for Google Vertex AI Search retriever.

Return type:

dict[str, tuple[type, Any]]

get_output_fields()

Return output field definitions for Google Vertex AI Search retriever.

Return type:

dict[str, tuple[type, Any]]

instantiate()

Create a Google Vertex AI Search retriever from this configuration.

Returns:

Instantiated retriever ready for enterprise search.

Return type:

GoogleVertexAISearchRetriever

Raises:

• ImportError – If required packages are not available.

• ValueError – If GCP credentials or configuration is invalid.