As AI evolves towards autonomous agents, integrating real-time data becomes crucial. This session explores how StarTree’s support for the Model Context Protocol (MCP) and native vector auto-embedding empowers AI agents with live, structured data access. Attendees will gain insights into building scalable pipelines that combine streaming ingestion (Kafka), real-time analytics (Apache Pinot), and AI models. A live demonstration will showcase the integration of Kafka → Pinot → AI agent, highlighting the architecture that enables real-time decision-making with contextually relevant information. This talk emphasizes practical takeaways for architecting systems that seamlessly blend AI and analytics technologies.
Over the last couple of years, the capabilities of large language models (LLMs) and Generative AI have evolved at a breakneck pace, with enterprise adoption accelerating rapidly. These technologies are not only driving automation but are fundamentally challenging traditional approaches to how data is engineered, managed, and consumed.
We are entering a new paradigm—one where emerging ways of working are redefining what’s possible across the enterprise data ecosystem. Join us for this Keynote as we explore what this transformation means for data engineers and business leaders. We’ll examine real-world use cases, the evolving skill sets needed to thrive, and how teams can adapt to stay ahead in a rapidly shifting landscape.
Your CEO has demanded that you leverage Generative AI technologies to lower costs and find even more insights out of your existing data sources. Every webinar you attend and every blog post you read worries you about problems such as hallucinations and gives you nightmares as you think about security, privacy, compliance, and governance. You imagine your entire workforce needs to be reshaped, and all of the time, effort, and money you’ve spent up until now, has netted you a pile of technical debt.
Worse yet, because senior leadership sees Gen AI as a bit of “magic”, you don’t have clearly-defined and measurable business objectives. You are surely going to fail!! Time to take a deep breath…
If there is hope for successfully adopting Gen AI technologies into your enterprise… it is in the data platforms you already have (or at least, are already available).
This talk explains the foundational problems of Gen AI applicability for enterprises are focused on data access, collaboration, and governance. Data platforms that tackle these concerns exist today and are likely already deployed, or upgradeable, in your IT arsenal. No need to throw out what you have already built; it is time to architect around your solid foundation as you charge bravely into a new adventure
The current technological landscape underscores a significant convergence, where functionalities traditionally associated with separate domains of data management and AI are now unified within intelligent data platforms. This integration marks a fundamental shift in perspective, moving away from the concept of AI as merely an application layer to AI-integrated data infrastructure. This amalgamation facilitates a more streamlined and effective utilization of data for AI initiatives, fastens the pace of iteration in model development and ensures a tighter alignment between an organization’s overall data strategy and AI vision. Furthermore, the emphasis placed on the delivery of “actionable insights” and the facilitation of “powerful decision-making” highlights the primary business-oriented purpose of these platforms. The evolution of data platforms reflects an imperative to not only manage the exponentially growing data volume but also derive intelligence that can inform and drive strategic decisions. Intelligent data platforms are engineered explicitly with this objective at their core, leveraging AI to automate insight and offer recommendations for accelerated organizational innovation.
The data landscape is transforming with generative AI driving innovations in pipeline architecture, real-time stream processing, and modern ETL practices. Emphasis on data trust, quality, and scalable monetization, alongside advancements in schema design and access layers, is unlocking new opportunities and efficiencies for businesses.
In an era where financial decisions hinge on millisecond precision, delivering real-time price streams to end-users is no longer a luxury — it’s a necessity. This talk explores modern best practices in building low-latency, highly reliable real-time data pipelines using technologies like WebSockets and NATS. We’ll dive into the architectural patterns that enable microservices to stream live pricing data seamlessly, and how tracking end-to-end latencies — from client browsers to backend systems — is both crucial and challenging.
Learn how these real-time streams not only power live charts and dashboards but also feed AI systems that detect anomalies, forecast trends, and trigger intelligent actions. If you’re building platforms where speed, scale, and accuracy are non-negotiable, this session will offer practical insights and proven techniques to get it right.
In today’s data-driven world, context is everything. Without context, raw data can often lead to inaccurate insights, misinterpretations, and missed opportunities. For data engineers, this means building platforms that enrich data with relevant business context and provide strategic insights.
Join this session to explore how modern, context-centric platforms help businesses stay ahead in the data-driven landscape.
As organizations race to adopt Generative AI, a strong and scalable data foundation is essential. This session explores how building a FAIR-aligned Data Marketplace has transformed data access, quality, and governance- paving the way for AI-driven innovation. Learn how a persona-driven, self-service platform powered by Snowflake, Power BI, and advanced metadata cataloging accelerated decision-making, eliminated redundancy, and boosted adoption by 25%. Key innovations include a real-time Data Quality Index, seamless integration for both technical and non-technical users, and smart search capabilities using LLMs. Discover how strategic leadership, automation, and a culture of collaboration are driving enterprise-wide data democratization- preparing the organization to fully harness the power of Generative AI.
In modern systems, real-time insights aren’t a luxury—they’re a requirement. Whether you’re debugging distributed systems, tracking financial transactions, or analyzing user behavior, sub-second query latency can be the difference between reacting and proactively optimizing.
This talk dives into the technical foundations that make real-time analytics possible at scale, using ClickHouse as the case study. We’ll explore the architectural underpinnings of its high-performance columnar engine, including vectorized execution, late materialization, and how it handles time series and semi-structured data like JSON with minimal overhead.
Through real-world use cases—from high-throughput log ingestion in observability stacks, to petabyte-scale analytics for adtech, fintech, and user personalization—you’ll see how engineering teams are designing for low latency without sacrificing flexibility or scale.
If you’re a data engineer, backend developer, or platform architect working with high-velocity data, this session will give you a deeper understanding of how to build infrastructure that can keep up.
