Is your AI evaluation process holding back your system's true potential? Many organizations struggle with improving GenAI quality because they don't know how to measure it effectively. This research session covers the principles of GenAI evaluation, offers a framework for measuring what truly matters, and demonstrates implementation using Databricks.Key Takeaways:-Practical approaches for establishing reliable metrics for subjective evaluations-Techniques for calibrating LLM judges to enable cost-effective, scalable assessment-Actionable frameworks for evaluation systems that evolve with your AI capabilitiesWhether you're developing models, implementing AI solutions, or leading technical teams, this session will equip you to define meaningful quality metrics for your specific use cases and build evaluation systems that expose what's working and what isn't, transforming AI guesswork into measurable success.

Taxonomy generation is a challenge across industries such as retail, manufacturing and e-commerce. Incomplete or inconsistent taxonomies can lead to fragmented data insights, missed monetization opportunities and stalled revenue growth. In this session, we will explore a modern approach to solving this problem by leveraging Databricks platform to build a scalable compound AI architecture for automated taxonomy generation. The first half of the session will walk you through the business significance and implications of taxonomy, followed by a technical deep dive in building an architecture for taxonomy implementation on the Databricks platform using a compound AI architecture. We will walk attendees through the anatomy of taxonomy generation, showcasing an innovative solution that combines multimodal and text-based LLMs, internal data sources and external API calls. This ensemble approach ensures more accurate, comprehensive and adaptable taxonomies that align with business needs.

In enterprise AI, Evaluation-Driven Development (EDD) ensures reliable, efficient systems by embedding continuous assessment and improvement into the AI development lifecycle. High-quality evaluation datasets are created using techniques like document analysis, synthetic data generation via Mosaic AI’s synthetic data generation API, SME validation, and relevance filtering, reducing manual effort and accelerating workflows. EDD focuses on metrics such as context relevance, groundedness, and response accuracy to identify and address issues like retrieval errors or model limitations. Custom LLM judges, tailored to domain-specific needs like PII detection or tone assessment, enhance evaluations. By leveraging tools like Mosaic AI Agent Framework and Agent Evaluation, MLflow, EDD automates data tracking, streamlines workflows, and quantifies improvements, transforming AI development for delivering scalable, high-performing systems that drive measurable organizational value.

LLM agents often drift into failure when prompts, retrieval, external data, and policies interact in unpredictable ways. This technical session introduces a repeatable, metric-driven framework for detecting, diagnosing, and correcting these undesirable behaviors in agentic systems at production scale. We demonstrate how to instrument the agent loop with fine-grained signals—tool-selection quality, error rates, action progression, latency, and domain-specific metrics—and send them into an evaluation layer (e.g. Galileo). This telemetry enables a virtuous cycle of system improvement. We present a practical example of a stock-trading system and show how brittle retrieval and faulty business logic cause undesirable behavior. We refactor prompts, adjust the retrieval pipeline—verifying recovery through improved metrics. Attendees will learn how to: add observability with minimal code change, pinpoint root causes via tracing, and drive continuous, metric-validated improvement.

While Databricks powers your data lakehouse, DataHub delivers the critical context layer connecting your entire ecosystem. We'll demonstrate how DataHub extends Unity Catalog to provide comprehensive metadata intelligence across platforms. DataHub's real-time platform:Cut AI model time-to-market with our unified REST and GraphQL APIs that ensure models train on reliable and compliant data from across platforms, with complete lineage trackingDecrease data incidents by 60% using our event-driven architecture that instantly propagates changes across systems*Transform data discovery from days to minutes with AI-powered search and natural language interfaces.Leaders use DataHub to transform Databricks data into integrated insights that drive business value. See our demo of syncback technology—detecting sensitive data and enforcing Databricks access controls automatically—plus our AI assistant that enhances' LLMs with cross-platform metadata.

Imagine performing complex regulatory checks in minutes instead of days. We made this a reality using GenAI on the Databricks Data Intelligence Platform. Join us for a deep dive into our journey from POC to a production-ready AI audit tool. Discover how we automated thousands of legal requirement checks in annual reports with remarkable speed and accuracy. Learn our blueprint for: High-Performance AI: Building a scalable, >90% accurate AI system with an optimized RAG pipeline that auditors praise. Robust Productionization: Achieving secure, governed deployment using Unity Catalog, MLflow, LLM-based evaluation, and MLOps best practices. This session provides actionable insights for deploying impactful, compliant GenAI in the enterprise.

Dive into the latest Llama 4 models. See for yourself how to unleash the power of Llama models and achieve next level performance with our curated set of practical tools, techniques and recipes. Join us as we dive into the world of Llama models, exploring their capabilities, developer tools, and exciting use cases. Discover how these innovative models are transforming industries and improving performance in real-world applications.

The demand for data engineering keeps growing, but data teams are bored by repetitive tasks, stumped by growing complexity and endlessly harassed by an unrelenting need for speed. What if AI could take the heavy lifting off your hands? What if we make the move away from code-generation and into config-generation — how much more could we achieve? In this session, we’ll explore how AI is revolutionizing data engineering, turning pain points into innovation. Whether you’re grappling with manual schema generation or struggling to ensure data quality, this session offers practical solutions to help you work smarter, not harder. You’ll walk away with a good idea of where AI is going to disrupt the data engineering workload, some good tips around how to accelerate your own workflows and an impending sense of doom around the future of the industry!

Matei is a legend of open source: he started the Apache Spark project in 2009, co-founded Databricks, and worked on other widely used data and AI software, including MLflow, Delta Lake, and Dolly. His most recent research is about combining large language models (LLMs) with external data sources, such as search systems, and improving their efficiency and result quality. This will be a conversation coverering the latest and greatest of UC, Data Intelligence, AI Governance, and more.

Go beyond the user interface and explore the cutting-edge technology driving AI/BI Genie. This session breaks down the AI/BI Genie architecture, showcasing how LLMs, retrieval-augmented generation (RAG) and finely tuned knowledge bases work together to deliver fast, accurate responses. We’ll also explore how AI agents orchestrate workflows, optimize query performance and continuously refine their understanding. Ideal for those who want to geek out about the tech stack behind Genie, this session offers a rare look at the magic under the hood.

Generic LLM metrics are useless until it meets your business needs.In this session we will dive deep into creating bespoke custom state-of-the-art AI metrics that matters to you. Discuss best practices on LLM evaluation strategies, when to use LLM judge vs. statistical metrics and many more. Through a live demo using Mosaic AI Framework, we will showcase: How you can build your own custom AI metric tailored to your needs for your GenAI application Implement autonomous AI evaluation suite for complex, multi-agent systems Generate ground truth data at scale and production monitoring strategies Drawing from extensive experience on working with customers on real-world use cases, we will share actionable insights on building a robust AI evaluation framework By the end of this session, you'll be equipped to create AI solutions that are not only powerful but also relevant to your organizations needs. Join us to transform your AI strategy and make a tangible impact on your business!

This presentation explores how Databricks' Data Intelligence Platform supports the development and deployment of responsible AI in credit decisioning, ensuring fairness, transparency and regulatory compliance. Key areas include bias and fairness monitoring using Lakehouse Monitoring to track demographic metrics and automated alerts for fairness thresholds. Transparency and explainability are enhanced through the Mosaic AI Agent Framework, SHAP values and LIME for feature importance auditing. Regulatory alignment is achieved via Unity Catalog for data lineage and AIBI dashboards for compliance monitoring. Additionally, LLM reliability and security are ensured through AI guardrails and synthetic datasets to validate model outputs and prevent discriminatory patterns. The platform integrates real-time SME and user feedback via Databricks Apps and AI/BI Genie Space.

This session will explore how developers can easily select, extract, filter, and control data pre-ingestion to accelerate safe AI. Learn how the Securiti and Databricks partnership empowers Databricks users by providing the critical foundation for unlocking scalability and accelerating trustworthy AI development and adoption.Key Takeaways:● Understand how to leverage data intelligence to establish a foundation for frameworks like OWASP top 10 for LLM’s, NIST AI RMF and Gartner’s TRiSM.● Learn how automated data curation and synching address specific risks while accelerating AI development in Databricks.● Discover how leading organizations are able to apply robust access controls across vast swaths of mostly unstructured data● Learn how to maintain data provenance and control as data is moved and transformed through complex pipelines in the Databricks platform.

Join us for this session on how to build AI finance agents with Databricks and LangChain. This session introduces a powerful approach to building AI agents by combining a modular framework that integrates LangChain, retrieval-augmented generation (RAG), and Databricks' unified data platform to build intelligent, adaptable finance agents. We’ll walk through the architecture and key components, including Databricks Unity Catalog, ML Flow, and Mosaic AI involved in building a system tailored for complex financial tasks like portfolio analysis, reporting automation, and real-time risk insights. We’ll also showcase a demo of one such agent in action - a Financial Analyst Agent. This agent emulates the expertise of a seasoned data analyst, delivering in-depth analysis in seconds - eliminating the need to wait hours or days for manual reports. The solution provides organizations with 24/7 access to advanced data analysis, enabling faster, smarter decision-making.

Seismic shift Large Language Models are unleashing on data engineering, challenging traditional workflows. LLMs obliterate inefficiencies and redefine productivity. AI powerhouses automate complex tasks like documentation, code translation, and data model development with unprecedented speed and precision. Integrating LLMs into tools promises to reduce offshore dependency, fostering agile onshore innovation. Harnessing LLMs' full potential involves challenges, requiring deep dives into domain-specific data and strategic business alignment. Session will addresses deploying LLMs effectively, overcoming data management hurdles, and fostering collaboration between engineers and stakeholders. Join us to explore a future where LLMs redefine possibilities, inviting you to embrace AI-driven innovation and position your organization as a leader in data engineering.

As enterprises adopt AI and Large Language Models (LLMs), securing and governing these models - and the data used to train them - is essential. In this session, learn how Databricks Partner PointGuard AI helps organizations implement the Databricks AI Security Framework to manage AI-specific risks, ensuring security, compliance, and governance across the entire AI lifecycle. Then, discover how Obsidian Security provides a robust approach to AI security, enabling organizations to confidently scale AI applications.

This session will showcase Bosch’s journey in consolidating supply chain information using the Databricks platform. It will dive into how Databricks not only acts as the central data lakehouse but also integrates seamlessly with transformative components such as dbt and Large Language Models (LLMs). The talk will highlight best practices, architectural considerations, and the value of an interoperable platform in driving actionable insights and operational excellence across complex supply chain processes. Key Topics and Sections Introduction & Business Context Brief Overview of Bosch’s Supply Chain Challenges and the Need for a Consolidated Data Platform. Strategic Importance of Data-Driven Decision-Making in a Global Supply Chain Environment. Databricks as the Core Data Platform Integrating dbt for Transformation Leveraging LLM Models for Enhanced Insights

Our project demonstrates building enterprise AI systems cost-effectively, focusing on matching merchant descriptors to known businesses. Using fine-tuned LLMs and advanced search, we created a solution rivaling alternatives at minimal cost. The system works in three steps: A fine-tuned Llama 3 8B model parses merchant descriptors into standardized components. A hybrid search system uses these components to find candidate matches in our database. A Llama 3 70B model then evaluates top candidates, with an AI judge reviewing results for hallucination. We achieved a 400% latency improvement while maintaining accuracy and keeping costs low and each fine-tuning round cost hundreds of dollars. Through careful optimization and simple architecture for a balance between cost, speed and accuracy, we show that small teams with modest budgets can tackle complex problems effectively using this technology. We share key insights on prompt engineering, fine-tuning and cost and latency management.

Cognizant developed a GenAI-driven market intelligence chatbot for RJR using Dash UI. This chatbot leverages Databricks Vector Search for vector embeddings and semantic search, along with the DBRX-Instruct LLM model to provide accurate and contextually relevant responses to user queries. The implementation involved loading prepared metadata into a Databricks vector database using the GTE model to create vector embeddings, indexing these embeddings for efficient semantic search, and integrating the DBRX-Instruct LLM into the chat system with prompts to guide the LLM in understanding and responding to user queries. The chatbot also generated responses containing URL links to dashboards with requested numerical values, enhancing user experience and productivity by reducing report navigation and discovery time by 30%. This project stands out due to its innovative AI application, advanced reasoning techniques, user-friendly interface, and seamless integration with MicroStrategy.

In this session, we will delve into the creation of an infrastructure, CI/CD processes and monitoring systems that facilitate the responsible and efficient deployment of Large Language Models (LLMs) at Intermountain Healthcare. Using the "AI Inventory Agents" project as a case study, we will showcase how an LLM Agent can assist in effort and impact estimates, as well as provide insights into various AI products, both custom-built and third-party hosted. This includes their responsible AI certification status, development status and monitoring status (lights on, performance, drift, etc.). Attendees will learn how to build and customize their own LLMOps infrastructure to ensure seamless deployment and monitoring of LLMs, adhering to responsible AI practices.