In the age of Generative AI, knowledge bases are the backbone of intelligent systems, enabling them to deliver accurate and context-aware responses. But how do you ensure that these knowledge bases remain up-to-date and relevant in a rapidly changing world? Enter Apache Airflow, a robust orchestration tool that streamlines the automation of data workflows. This talk will explore how Airflow can be leveraged to manage and update AI knowledge bases across multiple data sources. We’ll dive into the architecture, demonstrate how Airflow enables efficient data extraction, transformation, and loading (ETL), and share insights on tackling challenges like data consistency, scheduling, and scalability. Whether you’re building your own AI-driven systems or looking to optimize existing workflows, this session will provide practical takeaways to make the most of Apache Airflow in orchestrating intelligent solutions.

As organizations increasingly rely on data-driven applications, managing the diverse tools, data, and teams involved can create challenges. Amazon SageMaker Unified Studio addresses this by providing an integrated, governed platform to orchestrate end-to-end data and AI/ML workflows. In this workshop, we’ll explore how to leverage Amazon SageMaker Unified Studio to build and deploy scalable Apache Airflow workflows that span the data and AI/ML lifecycle. We’ll walk through real-world examples showcasing how this AWS service brings together familiar Airflow capabilities with SageMaker’s data processing, model training, and inference features - all within a unified, collaborative workspace. Key topics covered: Authoring and scheduling Airflow DAGs in SageMaker Unified Studio Understanding how Apache Airflow powers workflow orchestration under the hood Leveraging SageMaker capabilities like Notebooks, Data Wrangler, and Models Implementing centralized governance and workflow monitoring Enhancing productivity through unified development environments Join us to transform your ML workflow experience from complex and fragmented to streamlined and efficient.

As data workloads grow in complexity, teams need seamless orchestration to manage pipelines across batch, streaming, and AI/ML workflows. Apache Airflow provides a flexible and open-source way to orchestrate Databricks’ entire platform, from SQL analytics with Materialized Views (MVs) and Streaming Tables (STs) to AI/ML model training and deployment. In this session, we’ll showcase how Airflow can automate and optimize Databricks workflows, reducing costs and improving performance for large-scale data processing. We’ll highlight how MVs and STs eliminate manual incremental logic, enable real-time ingestion, and enhance query performance—all while maintaining governance and flexibility. Additionally, we’ll demonstrate how Airflow simplifies ML model lifecycle management by integrating Databricks’ AI/ML capabilities into end-to-end data pipelines. Whether you’re a dbt user seeking better performance, a data engineer managing streaming pipelines, or an ML practitioner scaling AI workloads, this session will provide actionable insights on using Airflow and Databricks together to build efficient, cost-effective, and future-proof data platforms.

As data workloads grow in complexity, teams need seamless orchestration to manage pipelines across batch, streaming, and AI/ML workflows. Apache Airflow provides a flexible and open-source way to orchestrate Databricks’ entire platform, from SQL analytics with Materialized Views (MVs) and Streaming Tables (STs) to AI/ML model training and deployment. In this session, we’ll showcase how Airflow can automate and optimize Databricks workflows, reducing costs and improving performance for large-scale data processing. We’ll highlight how MVs and STs eliminate manual incremental logic, enable real-time ingestion, and enhance query performance—all while maintaining governance and flexibility. Additionally, we’ll demonstrate how Airflow simplifies ML model lifecycle management by integrating Databricks’ AI/ML capabilities into end-to-end data pipelines. Whether you’re a dbt user seeking better performance, a data engineer managing streaming pipelines, or an ML practitioner scaling AI workloads, this session will provide actionable insights on using Airflow and Databricks together to build efficient, cost-effective, and future-proof data platforms.

This talk explores EDB’s journey from siloed reporting to a unified data platform, powered by Airflow. We’ll delve into the architectural evolution, showcasing how Airflow orchestrates a diverse range of use cases, from Analytics Engineering to complex MLOps pipelines. Learn how EDB leverages Airflow and Cosmos to integrate dbt for robust data transformations, ensuring data quality and consistency. We’ll provide a detailed case study of our MLOps implementation, demonstrating how Airflow manages training, inference, and model monitoring pipelines for Azure Machine Learning models. Discover the design considerations driven by our internal data governance framework and gain insights into our future plans for AIOps integration with Airflow.

The journey from ML model development to production deployment and monitoring is often complex and fragmented. How can teams overcome the chaos of disparate tools and processes? This session dives into how Apache Airflow serves as a unifying force in MLOps. We’ll begin with a look at the broader MLOps trends observed by Google within the Airflow community, highlighting how Airflow is evolving to meet these challenges and showcasing diverse MLOps use cases – both current and future. Then, Priceline will present a deep-dive case study on their MLOps transformation. Learn how they leveraged Cloud Composer, Google Cloud’s managed Apache Airflow service, to orchestrate their entire ML pipeline end-to-end: ETL, data preprocessing, model building & training, Dockerization, Google Artifact Registry integration, deployment, model serving, and evaluation. Discover how using Cloud Composer on GCP enabled them to build a scalable, reliable, adaptable, and maintainable MLOps practice, moving decisively from chaos to coordination. Cloud Composer (Airflow) has served as a major backbone in transforming the whole ML experience in Priceline. Join us to learn how to harness Airflow, particularly within a managed environment like Cloud Composer, for robust MLOps workflows, drawing lessons from both industry trends and a concrete, successful implementation.

Airflow powers thousands of data and ML pipelines—but in the enterprise, these pipelines often need to interact with business-critical systems like ERPs, CRMs, and core banking platforms. In this demo-driven session we will connect Airflow with Control-M from BMC and showcase how Airflow can participate in end-to-end workflows that span not just data platforms but also transactional business applications. Session highlights Trigger Airflow DAGs based on business events (e.g., invoice approvals, trade settlements) Feed Airflow pipeline outputs into ERP systems (e.g., SAP) or CRMs (e.g., Salesforce) Orchestrate multi-platform workflows from cloud to mainframe with SLA enforcement, dependency management, and centralized control. Provide unified monitoring and auditing across data and application layers

Airflow powers thousands of data and ML pipelines—but in the enterprise, these pipelines often need to interact with business-critical systems like ERPs, CRMs, and core banking platforms. In this demo-driven session we will connect Airflow with Control-M from BMC and showcase how Airflow can participate in end-to-end workflows that span not just data platforms but also transactional business applications. Session highlights Trigger Airflow DAGs based on business events (e.g., invoice approvals, trade settlements) Feed Airflow pipeline outputs into ERP systems (e.g., SAP) or CRMs (e.g., Salesforce) Orchestrate multi-platform workflows from cloud to mainframe with SLA enforcement, dependency management, and centralized control. Provide unified monitoring and auditing across data and application layers

In today’s data-driven world, scalable ML infrastructure is mission-critical. As ML workloads grow, orchestration tools like Apache Airflow become essential for managing pipelines, training, deployment, and observability. In this talk, I’ll share lessons from building distributed ML systems across cloud platforms, including GPU-based training and AI-powered healthcare. We’ll cover patterns for scaling Airflow DAGs, integrating telemetry and auto-healing, and aligning cross-functional teams. Whether you’re launching your first pipeline or managing ML at scale, you’ll gain practical strategies to make Airflow the backbone of your ML infrastructure.

As the demand for data products grows, data engineering teams face mounting pressure to deliver more and even faster, often becoming bottlenecks. Astro IDE changes the game. Astro IDE is an AI-powered code editor built for Apache Airflow. It helps data teams go from idea to production in minutes—generating production-ready DAGs, enabling in-browser testing, and integrating directly with Git. In this session, see how Astro IDE accelerates DAG creation, debugging, and deployment so data engineering teams can deliver more, 10x faster.

As the demand for data products grows, data engineering teams face mounting pressure to deliver more and even faster, often becoming bottlenecks. Astro IDE changes the game. Astro IDE is an AI-powered code editor built for Apache Airflow. It helps data teams go from idea to production in minutes—generating production-ready DAGs, enabling in-browser testing, and integrating directly with Git. In this session, see how Astro IDE accelerates DAG creation, debugging, and deployment so data engineering teams can deliver more, 10x faster.

Vayu is a conversational copilot for Apache Airflow, developed at Prevalent AI to help data engineers manage, troubleshoot, and fix pipelines using natural language. Deployments often fail silently due to misconfigurations, missing connections, or runtime issues impossible to identify in unit tests. Vayu tackles these via a troubleshooting agent that inspects logs, metrics, configs, and runtime state to find root causes and suggest fixes saving engineers significant troubleshooting time. It can also apply approved fixes to DAG code and commit them to your version control system. Key Capabilities: Troubleshooting Agent: Inspects logs, configs, variables, and connections to find root causes and suggest fixes. Pipeline Mechanic Agent: Suggests code-level fixes e.g., missing connections or bad imports and, once approved, commits them to version control. DAG Manager Agent: Understands DAG logic, suggests improvements, and can trigger DAGs conversationally. Architecture: Built with open-source tools including Google ADK as the orchestration layer and a custom Airflow MCP server based on the FastMCP framework. LLMs never access Airflow directly. The full codebase will be open-sourced.

On March 13th, 2025, Amazon Web Services announced General Availability of Amazon SageMaker Unified Studio, bringing together AWS machine learning and analytics capabilities. At the heart of this next generation of Amazon SageMaker sits Apache Airflow. All SageMaker Unified Studio users have a personal, open-source Airflow deployment, running alongside their Jupyter notebook, enabling those users to easily develop Airflow DAGs that have unified access to all of their data. In this talk, I will go into details around the motivations for choosing Airflow for this capability, the challenges with incorporating Airflow into such a large and diverse experience, the key role that open-source plays, how we’re leveraging GenAI to make that open source development experience better, and the goals for the future of Airflow in SageMaker Unified Studio. Attendees will leave with a better understanding of the considerations they need to make when choosing Airflow as a component of their enterprise project, and a greater appreciation of how Airflow can power advanced capabilities.

A practical demo and step-by-step guide on how to use AI to speed up and improve tasks.

A theoretical talk on the impact of AI on work, life and society.

A technical demo breaking down the process for building a product using AI with real-life learnings and insights.

The enterprise adoption of AI agents is accelerating, but significant challenges remain in making them truly reliable and effective. While coding assistants and customer service agents are already delivering value, more complex document-based workflows require sophisticated architectures and data processing capabilities. How do you design agent systems that can handle the complexity of enterprise documents with their tables, charts, and unstructured information? What's the right balance between general reasoning capabilities and constrained architectures for specific business tasks? Should you centralize your agent infrastructure or purchase vertical solutions for each department? The answers lie in understanding the fundamental trade-offs between flexibility, reliability, and the specific needs of your organization. Jerry Liu is the CEO and Co-founder at LlamaIndex, the AI agents platform for automating document workflows. Previously, he led the ML monitoring team at Robust Intelligence, did self-driving AI research at Uber ATG, and worked on recommendation systems at Quora. In the episode, Richie and Jerry explore the readiness of AI agents for enterprise use, the challenges developers face in building these agents, the importance of document processing and data structuring, the evolving landscape of AI agent frameworks like LlamaIndex, and much more. Links Mentioned in the Show: LlamaIndexLlamaIndex Production Ready Framework For LLM AgentsTutorial: Model Context Protocol (MCP)Connect with JerryCourse: Retrieval Augmented Generation (RAG) with LangChainRelated Episode: RAG 2.0 and The New Era of RAG Agents with Douwe Kiela, CEO at Contextual AI & Adjunct Professor at Stanford UniversityRewatch RADAR AI  New to DataCamp? Learn on the go using the DataCamp mobile appEmpower your business with world-class data and AI skills with DataCamp for business

Kasriel Kay, leading data democratization at Velotix, joined Yuliia and Dumke to challenge conventional wisdom about data governance and catalogs. Kasriel argues that data catalogs provide visibility but fail to deliver business value, comparing them to "buying JIRA and expecting agile practices." He advocates for shifting from restrictive data governance to data enablement through policy-based access control that considers user attributes, data sensitivity, and business context. Kasriel explains how AI-driven policy engines can learn from organizational behavior to automatically grant appropriate data access while maintaining compliance, ultimately reducing time-to-insight and unlocking missed business opportunities.

Summary In this episode of the Data Engineering Podcast Arun Joseph talks about developing and implementing agent platforms to empower businesses with agentic capabilities. From leading AI engineering at Deutsche Telekom to his current entrepreneurial venture focused on multi-agent systems, Arun shares insights on building agentic systems at an organizational scale, highlighting the importance of robust models, data connectivity, and orchestration loops. Listen in as he discusses the challenges of managing data context and cost in large-scale agent systems, the need for a unified context management platform to prevent data silos, and the potential for open-source projects like LMOS to provide a foundational substrate for agentic use cases that can transform enterprise architectures by enabling more efficient data management and decision-making processes.

Announcements Hello and welcome to the Data Engineering Podcast, the show about modern data managementData migrations are brutal. They drag on for months—sometimes years—burning through resources and crushing team morale. Datafold's AI-powered Migration Agent changes all that. Their unique combination of AI code translation and automated data validation has helped companies complete migrations up to 10 times faster than manual approaches. And they're so confident in their solution, they'll actually guarantee your timeline in writing. Ready to turn your year-long migration into weeks? Visit dataengineeringpodcast.com/datafold today for the details. This episode is brought to you by Coresignal, your go-to source for high-quality public web data to power best-in-class AI products. Instead of spending time collecting, cleaning, and enriching data in-house, use ready-made multi-source B2B data that can be smoothly integrated into your systems via APIs or as datasets. With over 3 billion data records from 15+ online sources, Coresignal delivers high-quality data on companies, employees, and jobs. It is powering decision-making for more than 700 companies across AI, investment, HR tech, sales tech, and market intelligence industries. A founding member of the Ethical Web Data Collection Initiative, Coresignal stands out not only for its data quality but also for its commitment to responsible data collection practices. Recognized as the top data provider by Datarade for two consecutive years, Coresignal is the go-to partner for those who need fresh, accurate, and ethically sourced B2B data at scale. Discover how Coresignal's data can enhance your AI platforms. Visit dataengineeringpodcast.com/coresignal to start your free 14-day trial. Your host is Tobias Macey and today I'm interviewing Arun Joseph about building an agent platform to empower the business to adopt agentic capabilitiesInterview IntroductionHow did you get involved in the area of data management?Can you start by giving an overview of how Deutsche Telekom has been approaching applications of generative AI?What are the key challenges that have slowed adoption/implementation?Enabling non-engineering teams to define and manage AI agents in production is a challenging goal. From a data engineering perspective, what does the abstraction layer for these teams look like? How do you manage the underlying data pipelines, versioning of agents, and monitoring of these user-defined agents?What was your process for developing the architecture and interfaces for what ultimately became the LMOS?How do the principles of operatings systems help with managing the abstractions and composability of the framework?Can you describe the overall architecture of the LMOS?What does a typical workflow look like for someone who wants to build a new agent use case?How do you handle data discovery and embedding generation to avoid unnecessary duplication of processing?With your focus on openness and local control, how do you see your work complementing projects like OumiWhat are the most interesting, innovative, or unexpected ways that you have seen LMOS used?What are the most interesting, unexpected, or challenging lessons that you have learned while working on LMOS?When is LMOS the wrong choice?What do you have planned for the future of LMOS and MASAIC?Contact Info LinkedInParting Question From your perspective, what is the biggest gap in the tooling or technology for data management today?Closing Announcements Thank you for listening! Don't forget to check out our other shows. Podcast.init covers the Python language, its community, and the innovative ways it is being used. The AI Engineering Podcast is your guide to the fast-moving world of building AI systems.Visit the site to subscribe to the show, sign up for the mailing list, and read the show notes.If you've learned something or tried out a project from the show then tell us about it! Email [email protected] with your story.Links LMOSDeutsche TelekomMASAICOpenAI Agents SDKRAG == Retrieval Augmented GenerationLangChainMarvin MinskyVector DatabaseMCP == Model Context ProtocolA2A (Agent to Agent) ProtocolQdrantLlamaIndexDVC == Data Version ControlKubernetesKotlinIstioXerox PARC)OODA (Observe, Orient, Decide, Act) LoopThe intro and outro music is from The Hug by The Freak Fandango Orchestra / CC BY-SA

We published our midyear outlook this week, highlighting expectations for a significant stagflationary shift in 2H25 and our limited confidence in forecasting the magnitude of this shift. We expect global GDP growth to fall well below potential, while the US tariff shock is expected to increase overall global inflation. Despite the uncertainty, we are observing a shift in market signals away from the narrative of weaker growth and rising inflation. One key message from the markets is a growing conviction in the likelihood of near-term Fed easing. We see the risk of a September easing as high, contingent upon further evidence of softening in the US labor markets.

Speakers:

Bruce Kasman

Joseph Lupton

This podcast was recorded on 27 June 2025.

This communication is provided for information purposes only. Institutional clients please visit www.jpmm.com/research/disclosures for important disclosures. © 2025 JPMorgan Chase & Co. All rights reserved. This material or any portion hereof may not be reprinted, sold or redistributed without the written consent of J.P. Morgan. It is strictly prohibited to use or share without prior written consent from J.P. Morgan any research material received from J.P. Morgan or an authorized third-party (“J.P. Morgan Data”) in any third-party artificial intelligence (“AI”) systems or models when such J.P. Morgan Data is accessible by a third-party. It is permissible to use J.P. Morgan Data for internal business purposes only in an AI system or model that protects the confidentiality of J.P. Morgan Data so as to prevent any and all access to or use of such J.P. Morgan Data by any third-party.