Druva Intros MetaGraph to Advance Agentic Data Security

Druva has introduced Dru MetaGraph, a secure, tenant-specific, graph-powered metadata layer to power real-time data intelligence. This move from the data security provider builds directly on its introduction last month of DruAI Agents, an AI-driven expansion of its DruAI platform, representing the company’s push into agentic AI systems for cyber resilience.

Unlocking Backup Metadata with MetaGraph

Dru MetaGraph aggregates all backup metadata — including file attributes, permissions, and identity information — into a secure, graph-based layer. It makes this data queryable and immediately usable for AI within the Druva platform, eliminating the need to extract, transform, and load metadata into external systems before analysis. According to Druva, this turns backup from a passive system of record into an active intelligence layer for security, compliance, and operational decision-making.

CEO and Co-founder Jaspreet Singh said, “Backup has always been an untapped goldmine of intelligence, and Druva has steadily unlocked that potential with our growing family of AI agents. With Dru MetaGraph and the new DruAI agents, we’re taking the next step and giving every team the ability to query and act on their metadata.”

The MetaGraph’s architecture enforces tenant-level isolation with end-to-end encryption and supports private retrieval-augmented generation (RAG) and isolated large language models (LLMs) to keep customer data secure while enabling contextual, AI-driven analysis.

Expanding the DruAI Fleet of Agents

Coinciding with the MetaGraph launch, Druva introduced two new DruAI Agents: the Insights Agent and the Lifecycle Agent. These join the initial agents announced in August — Data Agents, Help Agents, and Action Agents — which were built on Amazon Bedrock AgentCore within AWS infrastructure to interpret user intent, analyze data, and take actions.

The new agents are designed to directly leverage the MetaGraph:

• Insights Agent: Brings real-time intelligence into dashboards and reports, distilling complex data into prioritized insights. It highlights the most important risks, anomalies, and trends and recommends next steps, reducing the need to sift through numerous charts or alerts.
• Lifecycle Agent: Enables organizations to analyze metadata across the entire data lifecycle, from creation to archival. Using natural language queries, it can identify stale or non-compliant data, locate orphaned accounts, and enforce retention policies at scale. It also supports turning queries into ongoing visualizations and dashboards for continual monitoring.

These agents work alongside the earlier DruAI Agents, which include:

• Data Agents: Surface key signals and trends from telemetry, risk indicators, and historical data.
• Help Agents: Guide users by troubleshooting problems, investigating incidents, and recommending next steps.
• Action Agents: Execute tasks such as recovering workloads, creating protection policies, or adjusting retention settings to optimize cost.