PuppyGraph Raises $10M to Help AI Agents Query Enterprise Data Across 3,000 Tables in Under a Second

PuppyGraph Secures $10M Series A Funding to Scale Graph Query Engine for AI Agents

PuppyGraph announced it has raised $10 million in Series A funding led by Particle Future and defy.vc to expand its enterprise go-to-market operation and engineering capabilities2

2

. This marks the second time defy.vc has led a round for the Santa Clara-based startup, following a $5 million seed round in November 2024, bringing total investor capital to $15 million2

2

. The funding coincides with the general release of PuppyGraph 1.0, which introduces an AI-assisted graph schema builder and Model Context Protocol integrations designed to help AI agents query enterprise data more efficiently across thousands of database tables.

The company's annual recurring revenue grew sevenfold over the past year, though no specific dollar figure was disclosed2

2

. PuppyGraph's customer base includes Advanced Micro Devices Inc., Palo Alto Networks Inc., Coinbase Global Inc., Datadog Inc., and Trend Micro Inc.2

2

.

AI-Assisted Schema Builder Eliminates Manual Data Mapping

PuppyGraph 1.0's AI-assisted graph schema builder automatically maps existing tables into graph entities, relationships, and properties, allowing developers to review and approve the resulting schema rather than manually defining how disparate datasets connect1

1

. These schemas provide AI agents with a predefined map of how information across different tables and systems connects, simplifying query generation and reducing errors and retries1

1

.

The engine maps existing tables into a native Labeled Property Graph model, representing data entities as nodes, their relationships as edges, and their attributes as properties, before executing graph queries across underlying data sources through its distributed compute cluster1

1

. Xinyu Liu, a staff software engineer at Coinbase, noted that because the exchange connects directly to tables already in Snowflake and Databricks, "standing up a new graph is a schema definition, not a data migration"2

2

.

Model Context Protocol Integration Simplifies Agent Deployment

PuppyGraph 1.0 adds Model Context Protocol integrations that allow enterprise teams to connect PuppyGraph to AI agents and tools supporting the protocol1

1

. Through these integrations, supported agents can access PuppyGraph's graph query capabilities as part of agentic workflows without requiring developers to build custom integrations for each agent1

1

. The company says AI agents running on PuppyGraph can work across 3,000 enterprise tables and get graph query results back in under a second2

2

.

Devin Pratt, research director at IDC, explained that the ability to automatically generate graph schemas from existing tables and expose graph-query capabilities to compatible AI agents through MCP can reduce the engineering effort involved in connecting AI agents to enterprise data1

1

. "Developers model entities and relationships once, then query connected data with graph languages instead of writing and maintaining long chains of joins," Pratt said1

1

.

Query Validation Against Predefined Ontologies Reduces Errors

To improve query accuracy, PuppyGraph's query engine validates agent-generated queries against the graph schema, rejecting undefined relationships and returning feedback that agents can use to correct errors1

1

. Any agent that references a relationship nobody set up gets its query bounced, and the error that comes back is written for the agent to act on2

2

.

Ashish Chaturvedi, executive research leader at HFS Research, identified this enforced ontology as a significant gain for agentic workflows. The fact that the engine validates agent-generated queries against predefined ontologies helps constrain agents to the business entities and relationships established by the enterprise, reducing the scope for agents to generate invalid joins or follow relationships that do not exist in the underlying data1

1

. This means agents are less likely to return unreliable results because of faulty joins, Chaturvedi added1

1

.

Distributed Architecture Addresses Query Latency and Scalability Challenges

The execution engine splits complex graph queries into smaller operations that run simultaneously across multiple compute nodes, rather than sequentially on a single machine, helping accelerate queries that traverse multiple relationships across large datasets1

1

. This distributed approach addresses scalability challenges that emerge when agent projects move beyond pilot stages.

PuppyGraph's distributed graph query engine lets enterprises query relational databases, data warehouses and data lakehouses as a unified graph without moving or duplicating data into a separate graph database1

1

. According to PuppyGraph, agent projects tend to hit trouble once they leave the pilot stage, with schema descriptions alone eating into an agent's context window, and queries running several relationships deep can stall until they time out2

2

.

Cost Reduction Through Eliminated Data Duplication and Reduced Compute Spending

Pratt noted that PuppyGraph's approach could help CIOs reduce compute costs on agent queries by limiting the amount of schema information agents need to process and the number of query-generation retries1

1

. Enterprises can reduce data-management costs and complexity by avoiding duplicate storage and the synchronization pipelines required to maintain a separate graph database, while keeping graph analysis within their existing governed data environment1

1

.

PuppyGraph's approach aligns with enterprises' efforts to consolidate data-management capabilities. IDC research shows that 96% of organizations are actively pursuing such consolidation, while most manage more than 10 database technologies1

1

. The high cost of preparing, storing and serving data is among the most-cited barriers to scaling generative and agentic AI1

1

.

Security Features Include Row-Level Permissions and Short-Lived Credentials

When PuppyGraph connects to Snowflake Inc. or Databricks Inc., it can obtain short-lived credentials tied to the user's identity, eliminating the need to hand over long-lived keys2

2

. Row-level permissions configured on the data still apply, and the company says it checks them at each hop of a traversal2

2

.

Weimo Liu, co-founder and CEO of PuppyGraph, stated: "The insights of enterprise data will soon be generated by agents. Users will expect to ask questions that cross departments and systems, beyond the workflows someone built into a dashboard"2

2

. Amy Yin, a partner at defy.vc, noted that while writing a query is now a job a user can hand to Anthropic's Claude, executing it across billions of relationships still "takes serious engineering"2

2

.

Developers can download a free version of PuppyGraph with a single Docker command and no sign-up form, and the company says it takes about 10 minutes to go from deployment to a first query2

2

.