Standard SQL. Open formats. Your infrastructure.
Search, graph traversal, and analytics in standard SQL.
-- Search, graph traversal, and analytics in one query
SELECT
e.name,
e.entity_type,
graph_neighbors(e.id, 2) AS related,
COUNT(*) OVER () AS total_matches
FROM entities e
WHERE
text_search(e.description, 'regulatory compliance')
AND e.risk_score > 0.7
ORDER BY e.risk_score DESC
LIMIT 20;Run pipelines as functions. Materialize views with AI logic.
-- Classify documents and generate embeddings inline
CREATE MATERIALIZED VIEW enriched_contracts AS
SELECT
c.id,
c.title,
classify(c.body, 'risk', 'compliance', 'standard') AS category,
embed(c.body) AS embedding,
summarize(c.body, 100) AS summary
FROM contracts c
WHERE c.status = 'active';LLMs, embeddings, classification, code execution, and reranking. All callable from SQL as pipeline functions.
Built on open standards
Parquet, Arrow, and standard SQL. Your data stays portable.
Export anytime in standard formats. No proprietary schemas.
Hosted by us, or deployed on your infrastructure. Compute and data stay where you need them.
Built for regulated industries
Controls in place, formal audit planned. Documentation available on request.
BAA available for healthcare customers. PHI handling protocols in place.
DPA available. Data residency options for EU customers.
Enterprise customers can specify where data lives and where queries execute.
Need to review our security posture? Request our security documentation
Explore the docs or talk to our engineering team about your use case.