PuppyGraph Graph Query Engine

PuppyGraph is a graph query engine with a Santa Clara, California address in its footer. A 2024 announcement states the product launched in March 2024 and that the company raised a 5 million USD seed round led by defy.vc. It lets teams query relational databases and lakehouses as a graph without moving data.

The engine runs on top of existing stores with no ETL and no separate graph storage, and supports SQL, Gremlin and Cypher. The site cites a 10-hop query over billions of edges in 2.26 seconds on a four-node cluster and petabyte-scale data with auto-sharding. Named customers include Coinbase, AMD, Netskope and Palo Alto Networks.

• Zero-ETL graph engine: maps tables to a graph without copying data.
• Multi-language queries: SQL, Gremlin and Cypher.
• Scale-out: auto-sharding for petabyte-scale data.
• Lakehouse partners: Databricks Unity Catalog, Amazon S3 Tables, BigQuery and ClickHouse.
• PuppyGraph AI: Graph builder and Query modes added in version 1.16.0.
• Editions: free Developer Edition and priced Enterprise Edition on AWS AMI or Docker.

• A Vietnamese bank running fraud-ring queries directly on its data lakehouse.
• A security operations centre linking assets, users and alerts as a graph for investigation.
• A retailer analysing customer-product relationships from existing warehouse tables.

• Zero ETL: graph queries run on existing relational and lakehouse data, avoiding a second copy.
• A free single-node Developer Edition and a 30-day Enterprise trial allow low-cost evaluation.
• Deployment in about 10 minutes through Docker.
• Supports SQL, Gremlin and Cypher, so existing skills carry over.

Teams want graph analytics for fraud, security and knowledge graphs without the cost of maintaining a separate graph database and ETL pipelines. Querying lakehouse data as a graph is an emerging pattern. Vietnamese banks and security operations teams already holding data in warehouses or lakehouses can test graph analysis without a migration.

Address in Santa Clara, California (per footer). Product launched March 2024; 5 million USD seed round led by defy.vc announced November 2024. SOC 2 badge shown on the site.