EDB has unveiled a suite of innovative features for its Postgres AI platform, introducing an agentic database and converged analytics capabilities. These enhancements are designed to enable enterprises to efficiently run AI agents alongside transactional workloads on a unified PostgreSQL foundation.
Governance Tools and AI Integration
The latest release also includes governance tools in preview, which strategically position control mechanisms at the data layer, rather than relying on separate systems. The agentic database and converged analytics products, including EDB PG AI for ClickHouse, are now generally available, reflecting a broader industry trend among database and analytics providers to integrate AI processing more closely with operational data. This integration allows businesses to connect live enterprise records with AI systems without the need to transfer sensitive information across various platforms.
EDB’s agentic database feature aims to transform Postgres into a self-managing system, capable of monitoring over 200 operational and performance metrics. The software can proactively identify issues, suggest changes, and, depending on established policies, apply fixes automatically. Users have the flexibility to determine whether actions should be automated, sent for human approval, or postponed until a designated maintenance window, with all actions meticulously recorded in an audit trail.
According to EDB, this platform consolidates relational, JSON, time-series, geospatial, and vector data through a single SQL interface, facilitating operational, analytical, and AI workloads without necessitating data fragmentation across different engines. The system is reported to accelerate database tuning processes by up to tenfold, reducing tasks that typically require a database administrator 60 to 90 minutes to mere minutes. Furthermore, this optimization can enhance application performance for end users by as much as eight times.
Kevin Dallas, CEO of EDB, articulated the company’s perspective on the evolving market landscape: “The industry spent a decade telling enterprises to move everything into the lake. That’s exactly backwards for agents. Agents act in the moment, on live data, under real rules. You don’t get speed, accuracy, or sovereignty by reaching into a cloud for a copy. You get it by bringing the intelligence to the data. That’s what we built. Your AI, your data, your rules, on infrastructure you own.”
Advancements in Analytics
In tandem with the database enhancements, EDB has broadened the analytics capabilities of its platform through what it describes as a zero-ETL architecture. This architecture ensures that operational and analytical data are continuously available for real-time analysis and large-scale warehousing within the same environment.
EDB PG AI for ClickHouse is now generally available, targeting real-time analysis of event and log data. Meanwhile, EDB PG AI for WarehousePG focuses on historical analysis and complex reporting at petabyte scale, with heavier workloads being offloaded to GPU-accelerated Spark. EDB claims that these converged analytics functions can achieve up to 30 times faster single-node query performance compared to legacy warehouses and cloud data platforms, with performance improvements reaching up to 99 times through GPU offloading. Customers may also experience up to 52% greater scaling efficiency for high-concurrency workloads and up to 58% lower total cost of ownership.
Max Romanenko, Chief Engineering Officer of EDB, emphasized the platform’s innovative approach: “Every other approach asks you to move your data to the intelligence. We did the opposite—we put the intelligence in the database, on infrastructure you own. It’s the database that runs itself, on your terms. That’s not a feature you bolt on. It’s the foundation.”
EDB highlighted the case of Kyobo Book Centre, one of South Korea’s largest booksellers, which revamped its analytics environment based on an on-premises WarehousePG foundation. The retailer anticipates substantial total cost of ownership savings while establishing a data platform for AI and vector-based services.
Vector Search and Retrieval
In conjunction with the launch, EDB underscored the significance of vector search and retrieval for AI agents. The platform integrates structured and unstructured data, analytics, and vector search within a single query layer, enabling agents to operate with authorized data without depending on a separate vector database.
Benchmarking conducted by McKnight Consulting Group indicated that EDB’s platform demonstrated lower query latency and higher retrieval accuracy compared to competing platforms, including Databricks and MongoDB. EDB reported findings of up to 99.4% lower query latency than Databricks and 93% lower than MongoDB, along with a Recall@10 score of 0.911. New writes are queryable in just 12 milliseconds, a significant improvement over the 3.8 seconds required by Databricks. This rapid response is crucial for workloads that depend on real-time data, making the combination of speed, freshness, and transactional consistency particularly suitable for autonomous agents acting on live enterprise records.
NTT East, one of Japan’s largest telecommunications carriers, is leveraging EDB PG AI for AI-driven network operations. This deployment involves generative AI agents that detect, analyze, and respond to network issues within a private environment, ensuring that operational data remains under the carrier’s control.
Governance Functionality
Another key aspect of the release is a governance feature currently in preview. EDB’s approach utilizes native Postgres roles and row-level security to manage agent access at the point of data querying, eliminating the need for a separate control plane. An agent’s identity, purpose, permissions, and organizational policies are integrated into a constrained query executed directly by Postgres, ensuring that agent actions are subject to the same controls and auditing as those of human users.
“Agents don’t act on copies. They act on the real thing—live, governed, right where it sits, with no separate system to secure and no lake to fall out of sync with,” Romanenko stated.
Built on open Postgres and open table formats, the platform can be deployed on-premises, in hybrid environments, or across cloud infrastructures. EDB’s partner ecosystem includes notable names such as Dell, IBM, Nvidia, Red Hat, and Supermicro. Unnikrishnan Rajagopal, worldwide director for ISV ecosystem, GSIs, and alliances at IBM, remarked, “IBM Power and EDB Postgres AI are empowering enterprises for the AI-native era by providing a secured, sovereign, and AI-ready infrastructure foundation. Together, we enable a resilient data ecosystem that supports data sovereignty.”
