feat(ee): turbopuffer ai vector search driver (#2402)

* feat(ai): add AI_VECTOR_DRIVER and turbopuffer configuration

* feat(ai): carry workspace and target space ids on vector lifecycle events

* feat(ai): add vector driver interface and turbopuffer request helpers

* feat(ai): add pgvector driver behind the vector driver interface

* refactor(ai): route vector reads and writes through the vector driver

* feat(ai): add turbopuffer vector driver

* feat(ai): rebuild turbopuffer namespaces when the embedding model changes

* feat(ai): warm the vector namespace cache on session start

* fix(ai): harden turbopuffer misconfiguration and reset failure paths

* fix(ai): collapse blank-line runs in extracted page text

* fix(ai): skip full re-embed when ai search is re-enabled within the delete grace window

* sync

* fix(ai): store real embedding dimensions instead of serialized vector length

* fix(ai): filter search hits by the page's current space at query time

* fix(ai): retry the page moved-to-space vector patch job

* sync

* feat(ai): pre-warm the vector namespace

* fix(ai): pass AI_VECTOR_DRIVER through the client build config
This commit is contained in:
Philip Okugbe
2026-08-17 21:18:16 +01:00
committed by GitHub
parent 911c1057d6
commit 232beda471
19 changed files with 171 additions and 8 deletions
@@ -310,6 +310,31 @@ export class EnvironmentService {
return val === 'true';
}
getAiVectorDriver(): string {
return this.configService
.get<string>('AI_VECTOR_DRIVER', 'pgvector')
.toLowerCase();
}
getTurbopufferApiKey(): string {
return this.configService.get<string>('TURBOPUFFER_API_KEY');
}
getTurbopufferRegion(): string {
return this.configService.get<string>('TURBOPUFFER_REGION');
}
getTurbopufferBaseUrl(): string {
return this.configService.get<string>('TURBOPUFFER_BASE_URL');
}
getTurbopufferNamespacePrefix(): string {
return this.configService.get<string>(
'TURBOPUFFER_NAMESPACE_PREFIX',
'docmost',
);
}
getOpenAiApiKey(): string {
return this.configService.get<string>('OPENAI_API_KEY');
}
@@ -5,6 +5,7 @@ import {
IsOptional,
IsString,
IsUrl,
Matches,
MinLength,
ValidateIf,
validateSync,
@@ -108,6 +109,41 @@ export class EnvironmentVariables {
@IsString()
AI_DRIVER: string;
@IsOptional()
@ValidateIf((obj) => obj.AI_VECTOR_DRIVER)
@IsIn(['pgvector', 'turbopuffer'])
@IsString()
AI_VECTOR_DRIVER: string;
@ValidateIf((obj) => obj.AI_VECTOR_DRIVER === 'turbopuffer')
@IsNotEmpty()
@IsString()
TURBOPUFFER_API_KEY: string;
@ValidateIf(
(obj) =>
obj.AI_VECTOR_DRIVER === 'turbopuffer' && !obj.TURBOPUFFER_BASE_URL,
)
@IsNotEmpty({
message:
'TURBOPUFFER_REGION is required when AI_VECTOR_DRIVER is turbopuffer, unless TURBOPUFFER_BASE_URL is set',
})
@IsString()
TURBOPUFFER_REGION: string;
@IsOptional()
@ValidateIf((obj) => obj.TURBOPUFFER_BASE_URL != '' && obj.TURBOPUFFER_BASE_URL != null)
@IsUrl({ protocols: ['http', 'https'], require_tld: false })
TURBOPUFFER_BASE_URL: string;
@IsOptional()
@IsString()
@Matches(/^[A-Za-z0-9\-_.]{1,90}$/, {
message:
'TURBOPUFFER_NAMESPACE_PREFIX may only contain letters, digits, dot, dash, underscore (max 90 chars)',
})
TURBOPUFFER_NAMESPACE_PREFIX: string;
@IsOptional()
@IsString()
AI_EMBEDDING_MODEL: string;
@@ -61,6 +61,7 @@ export enum QueueJob {
WORKSPACE_DELETED = 'workspace-deleted',
WORKSPACE_CREATE_EMBEDDINGS = 'workspace-create-embeddings',
WORKSPACE_DELETE_EMBEDDINGS = 'workspace-delete-embeddings',
WORKSPACE_RESET_EMBEDDINGS = 'workspace-reset-embeddings',
GENERATE_PAGE_EMBEDDINGS = 'generate-page-embeddings',
DELETE_PAGE_EMBEDDINGS = 'delete-page-embeddings',
@@ -49,6 +49,10 @@ export class StaticModule implements OnModuleInit {
: undefined,
POSTHOG_HOST: this.environmentService.getPostHogHost(),
POSTHOG_KEY: this.environmentService.getPostHogKey(),
AI_VECTOR_DRIVER:
this.environmentService.getAiVectorDriver() === 'turbopuffer'
? 'turbopuffer'
: undefined,
};
const windowScriptContent = `<script>window.CONFIG=${JSON.stringify(configString)};</script>`;