FAQ
What is Curiosity Workspace?
Curiosity Workspace is a product that combines a knowledge graph, a search engine, and AI capabilities (NLP, embeddings, LLM-driven workflows) into a single environment for building data-driven applications.
Is Curiosity Workspace a graph database or a search engine?
Both—plus AI. Curiosity Workspace includes a graph layer (nodes/edges), a search layer (text + vector retrieval), and AI features that integrate with both.
How do I ingest data?
Most teams start with connectors:
- map source records into node/edge schemas
- commit nodes and edges
- re-run safely using stable keys
See Data Integration → Connectors.
How do I make data searchable?
Search is explicit: you choose which types and fields to index for text search and which to index for embeddings.
See:
What’s the difference between vector search and hybrid search?
- Vector search uses embeddings to retrieve by meaning.
- Hybrid search combines keyword and vector retrieval to improve overall relevance.
See Hybrid Search.
How do I add custom business logic?
Use custom endpoints to implement business logic close to the data (graph + search), then call those endpoints from UIs and integrations.
See Custom Endpoints.
How should I think about permissions?
Permissions should be designed early in production projects. Search and AI workflows must be permission-aware so users only retrieve authorized content.