As AI becomes more agent-driven, data management must evolve just as quickly. AI agents need real-time data, trusted context, strong governance, and reliable orchestration to operate responsibly.
Traditional data architectures weren't built for autonomous systems that make decisions and act. Data platforms must now deliver the speed, intelligence, and control required to support agentic workflows at scale.
DBTA recently held a roundtable webinar, Agentic AI and the Next Era of Data Management, with industry experts who explored how agentic AI is shaping the next era of data management.
According to Aaron Qayumi, senior product marketing manager at Reltio, agentic action requires rich context. AI needs to understand a web of interconnected concepts to act accurately.
For an agentic data foundation, companies need:
- Trusted, agent-ready, unified and resolved data products
- Context-rich knowledge graph with entities, metadata, relationships
- Open and interoperable platform with MCP and API-first architecture
- Real-time, event-driven performance in milliseconds at scale
Reltio provides the system of context for the enterprise, Qayumi explained. The Reltio Intelligent Data Graph provides unified entities, relationships, and meaning.
Clive Bearman, senior director product marketing at Qlik, said traditional architectures weren't built for agents. Closing the AI readiness gap is the defining challenge of the agentic era.
According to Bearman, there are four pillars every data architecture must deliver for successful AI, these include:
- Speed: Real-time and low-latency streaming pipelines push data where decisions are made instantly, not where they were made last quarter.
- Trust: Quality and validation autonomous agents compound data errors at scale. Quality is now an operational risk management issue.
- Context: Semantic layer agents need enriched, meaningful data and not just raw fields. Metadata, glossaries, and semantic layer turn data into knowledge.
- Governance: Traceability and control. Autonomy demands accountability. Every decision must be explainable through your governance framework.
Data governance is not a burden, it is a superpower, Bearman said. It drives confidence in data by providing data lineage, role-based access, anomaly monitoring, and creating an audit trail.
Qlik’s Agentic Data Engineering can elevate your productivity through agentic AI, he explained.
Krishna Kumar, field CTO, data foundations at Informatica from Salesforce, said organizations are moving from automation to autonomy. However, there is a context gap. Context offers an accumulated understanding of your business. The outcome of emphasizing context is:
- More accurate and trustworthy Insights and AI decisions
- Hyper-personalized consistent customer experiences
- Adaptive Autonomation of business processes
- Increased efficiency and productivity
- Democratization - Scalable, governed AI innovation across the organization
Gerald Mann, VP of presales at Graphwise, explained the enterprise AI maturity model, the journey toward a 5-star AI enterprise. Critical business information is trapped in silos and difficult to find. Disconnected systems create data visibility and compliance challenges. Limited cross-functional visibility slows decision making. AI initiatives face high costs and slow ROI from processing raw data. Enterprise-wide AI operates on trusted context, maximizing business value.
For the full webinar, featuring a more in-depth discussion, Q&A, and more, you can view an archived version of the webinar here.