Enable KDB.AI vector search workflows for similarity, hybrid search, and time-series patterns.
npx skills add https://github.com/KxSystems/kx-skills --skill kdbai
Please help me install this Agent Skill. Skill: kdbai Source: https://github.com/KxSystems/kx-skills/tree/main/plugins/kdbai-knowledge/skills/kdbai Command: npx skills add https://github.com/KxSystems/kx-skills --skill kdbai
KDB.AI provides a scalable vector database and tooling to implement AI-ready similarity search, hybrid search, and time-series pattern matching across large datasets.
Run a basic KDB.AI vector search example using the Python client to retrieve top-k similar items from a sample dataset.
High-intent search queries and answers about installing and using this skill.
To build a RAG pipeline using vector search, you can use KDB.AI to store embeddings and retrieve top-k similar items. This provides a scalable vector database for similarity matching across large datasets.
Hybrid search handles reranking by combining dense vectors with sparse signals to improve relevance. This approach merges semantic similarity with keyword filters to refine search results.
Yes, you can perform time-series similarity search using dynamic time warping and time-series similarity support. This enables time-aware queries to match temporal patterns across large datasets.
You query a vector database using the Python client by managing kdbai_client-based queries, handling embeddings, defining table schemas, and utilizing REST endpoints to return top-k similar items.
KDB.AI supports GPU indexing for large-scale similarity search through dynamic indexing with CAGRA GPU indices. This ensures robust, production-grade operation for querying extensive datasets.
patricio0312revConfigure Pinecone, Chroma, pgvector, and Qdrant vector databases for semantic search.
edangx100Rank retrieved documents by similarity_score or distance with thresholds and top-k selection.
micropredictionCalculate real-time covariance for named time-varying series in Python.
matthew-johnsonPerform scalable vector similarity search and clustering across billions of embeddings.
