Day 1 | Hall 1 - Thought Leadership and Strategic Insights

In a world where AI can build pipelines and infer schemas, the role of the data engineer is undergoing a fundamental shift. The skills that once defined the profession are being automated and that raises an important question: where does a data engineer’s value truly lie?

Join this session to explore how a data engineer’s role is not being replaced, but rather elevated. Understand how you can move beyond manual execution and step into higher-impact roles, shaping business outcomes, driving adoption, and applying the human judgment that no model can match.

Every enterprise wants AI, and most have more raw material than they realize. Industry research by IDC shows that while over 80% of enterprise content is unstructured, less than 1% of it is currently ready for AI consumption. This gap is what we call the “Context Crisis”, which is the difference between data that’s stored and data that’s truly seen. Closing it is what turns models that guess into models that know, and agents that assist into agents that can be trusted with real decisions.

This keynote argues that AI reliability is a data engineering problem before it is ever a model problem. Drawing from real-world failure patterns, like dark data, broken lineage, missing metadata, unversioned documents, and more, we’ll explore how modern DataOps disciplines turn this crisis into a competitive edge. From intelligent ingestion and semantic layers to data products and governance as code, we’ll walk through the engineering blueprint that transforms fragmented documents and unversioned information into the contextual fuel agentic AI demands.

This session explores how generative AI is reshaping product discovery into a more intuitive, predictive, and user-centric experience.

Every enterprise is rushing to deploy AI. Few have fixed the foundation underneath it. Despite years of investment in data platforms, most enterprise AI systems still hallucinate, produce inconsistent outputs, and fail to reason reliably. The problem is not the model — it is a structural gap between how data platforms were built and what AI agents actually need. This session defines the AI Readiness Gap, diagnoses where enterprises are falling short across six critical dimensions and introduces the Knowledge Layer — the most overlooked component in modern data architecture that bridges raw data assets with AI agent cognition.

As organizations rush to layer AI on top of their data warehouses, most are discovering an uncomfortable truth. AI-driven analytics is only as trustworthy as the metadata underneath it. Without lineage, ownership, and semantic context, even the best models confidently produce wrong answers, and no one can tell until a dashboard lies to an executive.
In this talk, I’ll share how we built a data catalog that serves as the backbone of our AI-driven analysis infrastructure. We’ll walk through the pillars that make it work: end-to-end lineage from source systems to dashboards, blast-radius analysis that prevents silent breakage, documentation and accountability that AI agents can actually reason over, and a metrics layer that turns all of this into reliable, explainable answers.
You’ll leave with a concrete blueprint for why catalogs aren’t documentation tools, they’re infrastructure, and why getting this foundation right is the difference between AI that hallucinates and AI that works.

In this session, we’ll explore how large analytics enterprises can evolve from traditional data systems to AI-driven, modular architectures. The talk introduces five engineering pillars—Data Foundations, AI Agent Building (including persona-based “ask” capabilities), Agent Catalog & Monitoring, Internal Agent Marketplace, and Business Integration. Attendees will learn how data engineers can design semantic layers, build reusable AI agents for merchandising and customer analytics, and integrate them into live business workflows. The session highlights practical strategies for measuring impact, optimizing AI performance, and creating scalable solutions that drive measurable business outcomes.

Creating a single source of truth for customer data is no longer optional. This session will dive into the strategies, architectures, and governance models needed to integrate fragmented data systems into a unified ecosystem that powers seamless experiences and data-driven growth.

What does data governance look like when your biggest data consumer doesn’t sleep, doesn’t follow SOPs, and makes thousands of decisions every second? When a majority of your users bypass official AI tools entirely? When a single ungoverned prompt can leak training data, customer PII, or regulated information across your entire enterprise?
These aren’t hypothetical questions — they’re the operational reality of GenAI in 2026. And the governance frameworks most enterprises rely on were built for an era when data consumers were human, deterministic, and slow.

This talk presents DataGovOps: governance as code, dynamic access policies, AI-assisted classification, and end-to-end lineage from source through agent output. We’ll explore why legacy governance fails under agentic workloads, how to architect consent and access as programmable policy, the recursion problem of governing AI with AI, and practical patterns for staying compliant under regimes like GDPR & DPDPA — without strangling the velocity that makes AI worth deploying in the first place.

With tool sprawl, hidden costs, and integration nightmares, is the MDS era over? Panel can debate the rise of unified platforms vs. best-of-breed, and what the next architecture era looks like.