Summary In this crossover episode of the AI Engineering Podcast, host Tobias Macey interviews Brijesh Tripathi, CEO of Flex AI, about revolutionizing AI engineering by removing DevOps burdens through "workload as a service". Brijesh shares his expertise from leading AI/HPC architecture at Intel and deploying supercomputers like Aurora, highlighting how access friction and idle infrastructure slow progress. Join them as they discuss Flex AI's innovative approach to simplifying heterogeneous compute, standardizing on consistent Kubernetes layers, and abstracting inference across various accelerators, allowing teams to iterate faster without wrestling with drivers, libraries, or cloud-by-cloud differences. Brijesh also shares insights into Flex AI's strategies for lifting utilization, protecting real-time workloads, and spanning the full lifecycle from fine-tuning to autoscaled inference, all while keeping complexity at bay.

Pre-amble I hope you enjoy this cross-over episode of the AI Engineering Podcast, another show that I run to act as your guide to the fast-moving world of building scalable and maintainable AI systems. As generative AI models have grown more powerful and are being applied to a broader range of use cases, the lines between data and AI engineering are becoming increasingly blurry. The responsibilities of data teams are being extended into the realm of context engineering, as well as designing and supporting new infrastructure elements that serve the needs of agentic applications. This episode is an example of the types of work that are not easily categorized into one or the other camp.

Announcements Hello and welcome to the Data Engineering Podcast, the show about modern data managementData teams everywhere face the same problem: they're forcing ML models, streaming data, and real-time processing through orchestration tools built for simple ETL. The result? Inflexible infrastructure that can't adapt to different workloads. That's why Cash App and Cisco rely on Prefect. Cash App's fraud detection team got what they needed - flexible compute options, isolated environments for custom packages, and seamless data exchange between workflows. Each model runs on the right infrastructure, whether that's high-memory machines or distributed compute. Orchestration is the foundation that determines whether your data team ships or struggles. ETL, ML model training, AI Engineering, Streaming - Prefect runs it all from ingestion to activation in one platform. Whoop and 1Password also trust Prefect for their data operations. If these industry leaders use Prefect for critical workflows, see what it can do for you at dataengineeringpodcast.com/prefect.Data 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. Your host is Tobias Macey and today I'm interviewing Brijesh Tripathi about FlexAI, a platform offering a service-oriented abstraction for AI workloadsInterview IntroductionHow did you get involved in machine learning?Can you describe what FlexAI is and the story behind it?What are some examples of the ways that infrastructure challenges contribute to friction in developing and operating AI applications?How do those challenges contribute to issues when scaling new applications/businesses that are founded on AI?There are numerous managed services and deployable operational elements for operationalizing AI systems. What are some of the main pitfalls that teams need to be aware of when determining how much of that infrastructure to own themselves?Orchestration is a key element of managing the data and model lifecycles of these applications. How does your approach of "workload as a service" help to mitigate some of the complexities in the overall maintenance of that workload?Can you describe the design and architecture of the FlexAI platform?How has the implementation evolved from when you first started working on it?For someone who is going to build on top of FlexAI, what are the primary interfaces and concepts that they need to be aware of?Can you describe the workflow of going from problem to deployment for an AI workload using FlexAI?One of the perennial challenges of making a well-integrated platform is that there are inevitably pre-existing workloads that don't map cleanly onto the assumptions of the vendor. What are the affordances and escape hatches that you have built in to allow partial/incremental adoption of your service?What are the elements of AI workloads and applications that you are explicitly not trying to solve for?What are the most interesting, innovative, or unexpected ways that you have seen FlexAI used?What are the most interesting, unexpected, or challenging lessons that you have learned while working on FlexAI?When is FlexAI the wrong choice?What do you have planned for the future of FlexAI?Contact Info LinkedInParting Question From your perspective, what are the biggest gaps in tooling, technology, or training for AI systems today?Links Flex AIAurora Super ComputerCoreWeaveKubernetesCUDAROCmTensor Processing Unit (TPU)PyTorchTritonTrainiumASIC == Application Specific Integrated CircuitSOC == System On a ChipLoveableFlexAI BlueprintsTenstorrentThe intro and outro music is from The Hug by The Freak Fandango Orchestra / CC BY-SA

Summary In this episode of the AI Engineering Podcast Mark Brooker, VP and Distinguished Engineer at AWS, talks about how agentic workflows are transforming database usage and infrastructure design. He discusses the evolving role of data in AI systems, from traditional models to more modern approaches like vectors, RAG, and relational databases. Mark explains why agents require serverless, elastic, and operationally simple databases, and how AWS solutions like Aurora and DSQL address these needs with features such as rapid provisioning, automated patching, geodistribution, and spiky usage. The conversation covers topics including tool calling, improved model capabilities, state in agents versus stateless LLM calls, and the role of Lambda and AgentCore for long-running, session-isolated agents. Mark also touches on the shift from local MCP tools to secure, remote endpoints, the rise of object storage as a durable backplane, and the need for better identity and authorization models. The episode highlights real-world patterns like agent-driven SQL fuzzing and plan analysis, while identifying gaps in simplifying data access, hardening ops for autonomous systems, and evolving serverless database ergonomics to keep pace with agentic development.

Announcements Hello and welcome to the Data Engineering Podcast, the show about modern data managementData teams everywhere face the same problem: they're forcing ML models, streaming data, and real-time processing through orchestration tools built for simple ETL. The result? Inflexible infrastructure that can't adapt to different workloads. That's why Cash App and Cisco rely on Prefect. Cash App's fraud detection team got what they needed - flexible compute options, isolated environments for custom packages, and seamless data exchange between workflows. Each model runs on the right infrastructure, whether that's high-memory machines or distributed compute. Orchestration is the foundation that determines whether your data team ships or struggles. ETL, ML model training, AI Engineering, Streaming - Prefect runs it all from ingestion to activation in one platform. Whoop and 1Password also trust Prefect for their data operations. If these industry leaders use Prefect for critical workflows, see what it can do for you at dataengineeringpodcast.com/prefect.Data 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.Your host is Tobias Macey and today I'm interviewing Marc Brooker about the impact of agentic workflows on database usage patterns and how they change the architectural requirements for databasesInterview IntroductionHow did you get involved in the area of data management?Can you describe what the role of the database is in agentic workflows?There are numerous types of databases, with relational being the most prevalent. How does the type and purpose of an agent inform the type of database that should be used?Anecdotally I have heard about how agentic workloads have become the predominant "customers" of services like Neon and Fly.io. How would you characterize the different patterns of scale for agentic AI applications? (e.g. proliferation of agents, monolithic agents, multi-agent, etc.)What are some of the most significant impacts on workload and access patterns for data storage and retrieval that agents introduce?What are the categorical differences in that behavior as compared to programmatic/automated systems?You have spent a substantial amount of time on Lambda at AWS. Given that LLMs are effectively stateless, how does the added ephemerality of serverless functions impact design and performance considerations around having to "re-hydrate" context when interacting with agents?What are the most interesting, innovative, or unexpected ways that you have seen serverless and database systems used for agentic workloads?What are the most interesting, unexpected, or challenging lessons that you have learned while working on technologies that are supporting agentic applications?Contact Info BlogLinkedInParting 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 AWS Aurora DSQLAWS LambdaThree Tier ArchitectureVector DatabaseGraph DatabaseRelational DatabaseVector EmbeddingRAG == Retrieval Augmented GenerationAI Engineering Podcast EpisodeGraphRAGAI Engineering Podcast EpisodeLLM Tool CallingMCP == Model Context ProtocolA2A == Agent 2 Agent ProtocolAWS Bedrock AgentCoreStrandsLangChainKiroThe intro and outro music is from The Hug by The Freak Fandango Orchestra / CC BY-SA

Summary

PostGreSQL has become one of the most popular and widely used databases, and for good reason. The level of extensibility that it supports has allowed it to be used in virtually every environment. At Citus Data they have built an extension to support running it in a distributed fashion across large volumes of data with parallelized queries for improved performance. In this episode Ozgun Erdogan, the CTO of Citus, and Craig Kerstiens, Citus Product Manager, discuss how the company got started, the work that they are doing to scale out PostGreSQL, and how you can start using it in your environment.

Preamble

Hello and welcome to the Data Engineering Podcast, the show about modern data infrastructure When you’re ready to launch your next project you’ll need somewhere to deploy it. Check out Linode at dataengineeringpodcast.com/linode and get a $20 credit to try out their fast and reliable Linux virtual servers for running your data pipelines or trying out the tools you hear about on the show. Continuous delivery lets you get new features in front of your users as fast as possible without introducing bugs or breaking production and GoCD is the open source platform made by the people at Thoughtworks who wrote the book about it. Go to dataengineeringpodcast.com/gocd to download and launch it today. Enterprise add-ons and professional support are available for added peace of mind. Go to dataengineeringpodcast.com to subscribe to the show, sign up for the newsletter, read the show notes, and get in touch. You can help support the show by checking out the Patreon page which is linked from the site. To help other people find the show you can leave a review on iTunes, or Google Play Music, and tell your friends and co-workers Your host is Tobias Macey and today I’m interviewing Ozgun Erdogan and Craig Kerstiens about Citus, worry free PostGreSQL

Interview

Introduction How did you get involved in the area of data management? Can you describe what Citus is and how the project got started? Why did you start with Postgres vs. building something from the ground up? What was the reasoning behind converting Citus from a fork of PostGres to being an extension and releasing an open source version? How well does Citus work with other Postgres extensions, such as PostGIS, PipelineDB, or Timescale? How does Citus compare to options such as PostGres-XL or the Postgres compatible Aurora service from Amazon? How does Citus operate under the covers to enable clustering and replication across multiple hosts? What are the failure modes of Citus and how does it handle loss of nodes in the cluster? For someone who is interested in migrating to Citus, what is involved in getting it deployed and moving the data out of an existing system? How do the different options for leveraging Citus compare to each other and how do you determine which features to release or withhold in the open source version? Are there any use cases that Citus enables which would be impractical to attempt in native Postgres? What have been some of the most challenging aspects of building the Citus extension? What are the situations where you would advise against using Citus? What are some of the most interesting or impressive uses of Citus that you have seen? What are some of the features that you have planned for future releases of Citus?

Contact Info

Citus Data

citusdata.com @citusdata on Twitter citusdata on GitHub

Craig

Email Website @craigkerstiens on Twitter

Ozgun

Email ozgune on GitHub

Parting Question

From your perspective, what is the biggest gap in the tooling or technology for data management today?

Links

Citus Data PostGreSQL NoSQL Timescale SQL blog post PostGIS PostGreSQL Graph Database JSONB Data Type PipelineDB Timescale PostGres-XL Aurora PostGres Amazon RDS Streaming Replication CitusMX CTE (Common Table Expression) HipMunk Citus Sharding Blog Post Wal-e Wal-g Heap Analytics HyperLogLog C-Store

The intro and outro musi