Techsplainers by IBM

Techsplainers by IBM

Download on the App Store

Techsplainers by IBM episodes

  • What is streaming analytics?
    This episode of Techsplainers explores streaming analytics, the real-time analytics layer that helps organizations continuously ingest, process and analyze streaming data as it is generated. The episode explains how streaming analytics differs from traditional batch-based approaches by delivering insights with minimal latency, enabling businesses to act on events while they are still unfolding.
    Listeners are guided through why streaming analytics matters in modern business environments shaped by IoT, SaaS applications, financial transactions, social media and other continuous data sources. The discussion breaks down the four major components of a streaming analytics workflow—data ingestion, data processing and analysis, governance, and consumption—while also clarifying the subtle distinction between streaming analytics and real-time analytics.
    The episode highlights the growing connection between streaming analytics and AI, including support for AI agents, chatbots, recommendation engines, fraud detection models and predictive maintenance systems that depend on fresh, contextualized data. It also examines key benefits such as faster decision-making, better customer experiences, improved operational efficiency and stronger security, alongside the challenges of data integration, quality, governance, scalability and low-latency delivery. Major technologies such as Apache Kafka, Apache Flink, Spark Structured Streaming, Confluent and Iceberg are also introduced.
    Find more information at https://www.ibm.com/think/topics/streaming-analytics Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Ian Smalley
    10 min
  • What is event-driven architecture?
    This episode of Techsplainers explores event-driven architecture, a software design approach built around detecting, publishing and responding to events as they happen. The episode explains how EDA differs from more traditional architectures by focusing on real-time data in motion instead of static data stored for later analysis. It walks through the core building blocks of an event-driven system—event producers, brokers and consumers—and shows how those components work together to support fast, asynchronous communication across distributed environments.
    Listeners are introduced to key architectural models, including publish-subscribe and event streaming, along with the event processing patterns that power everything from immediate reactions to advanced pattern detection. The discussion also highlights why EDA is so valuable in cloud-native and microservices environments, where loose coupling improves scalability, resilience and flexibility. Real-world applications include suspicious activity detection, predictive maintenance, dynamic pricing and inventory optimization. The episode also covers the practical challenges teams face, such as event ordering, eventual consistency, observability and duplicate handling.
    Find more information at https://www.ibm.com/think/topics/event-driven-architectureFind more episodes https://www.ibm.biz/techsplainers-podcast
    Narrated by Ian Smalley
    8 min
  • What is Apache Kafka?
    This episode of Techsplainers explores Apache Kafka, the open-source event-streaming platform that has become a cornerstone of modern real-time data systems. The episode explains how Kafka helps organizations publish, store, process and consume streams of events with high throughput, low latency and strong reliability.
    Listeners are introduced to the concept of event streaming through familiar examples, from customer orders and website clicks to IoT sensor updates, before learning how Kafka differs from traditional message queues by retaining records for replay and independent consumption. The discussion then breaks down Kafka’s core architecture, including topics, partitions, brokers, offsets and replication, along with Kafka’s shift away from ZooKeeper toward KRaft for simpler cluster management.
    The episode also covers Kafka’s four primary APIs, its leading use cases in real-time data pipelines, streaming applications, microservices, cloud-native systems and IoT, and its broader ecosystem integrations with tools like Spark, Flink and Cassandra. Finally, it examines how Kafka compares with RabbitMQ and why Kafka is increasingly valuable in AI-driven systems that depend on live, continuously moving data.
    Find more information at https://www.ibm.com/think/topics/apache-kafka Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Ian Smalley
    10 min
  • What is a data streaming platform?
    This episode of Techsplainers explores data streaming platforms and how they help organizations continuously capture, process, analyze and deliver data in real time or near-real time. The episode explains why these platforms have become essential as businesses shift from slower batch-based workflows to real-time decision-making powered by live data from applications, databases, sensors, logs and digital interactions.
    Listeners are guided through the business context behind this shift, including the rise of big data, event-driven architectures and the need for faster analytics, AI responsiveness and operational agility. The episode also examines how data streaming platforms support AI use cases such as fraud detection, recommendation engines, predictive maintenance and AI agents that depend on current context.
    From there, the discussion breaks down the four core architectural layers of a data streaming platform—source and ingestion, processing, destination and serving, and governance and management—while highlighting key characteristics such as scalability, fault tolerance, low latency and high throughput. The episode also reviews major open-source and managed technologies, including Apache Kafka, Apache Flink, Apache Spark, Confluent, Amazon Kinesis, Google Cloud Dataflow and Azure Stream Analytics.
    Find more information at https://www.ibm.com/think/topics/data-streaming-platform Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Ian Smalley
    9 min
  • What is real-time data streaming?
    This episode of Techsplainers explores real-time data streaming and why it has become essential for modern enterprises that need immediate insight from fast-moving information. The episode explains how real-time streaming differs from traditional batch processing by handling data as it arrives, often within milliseconds, rather than waiting for scheduled runs.
    Listeners are guided through the main business benefits of this approach, including faster decision-making, improved operational efficiency, smarter data retention, decision-making, improved operational efficiency, smarter data retention, stronger risk management and more personalized customer experiences. The episode also highlights real-world applications across retail, banking, manufacturing and AI-driven systems that depend on current, continuously updated information.
    From there, the discussion breaks down the core components of streaming architecture—ingestion, processing and destination—and clarifies the close relationship between real-time data streaming and event streaming. It also reviews common technologies, such as Apache Kafka, Apache Flink and Spark Streaming, while addressing practical implementation challenges like cost, fault tolerance, observability, security and governance.
    Find more information at https://www.ibm.com/think/topics/real-time-data-streaming Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Ian Smalley
    8 min
  • What is crewAI?
    This episode of Techsplainers explores Crew AI, an open-source Python framework that enables the creation of collaborative AI teams with specialized roles and responsibilities. Developed by Joao Moura, Crew AI implements a manager-worker hierarchy where a coordinator agent delegates tasks to specialized workers based on their defined capabilities. We examine how this structured approach to collaboration allows AI teams to tackle complex, multifaceted projects by leveraging the complementary skills of different agents. The discussion details Crew AI's key components, including its role definition system, task delegation framework, structured collaboration protocol, tool integration capabilities, and memory management features. We highlight practical applications across content creation, product development, data analysis, and customer service, where role-based collaboration can deliver more comprehensive and nuanced results than single-agent approaches. While acknowledging current limitations in handling highly specialized tasks and the dependence on underlying language models, the episode emphasizes Crew AI's significance as a flexible, intuitive framework for organizing AI collaboration around clearly defined roles and responsibilities.

    Find more information at https://www.ibm.com/think/topics/crew-ai Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Amanda Downie
    7 min
  • What is LangChain?
    This episode of Techsplainers explores LangChain, an open-source framework that's revolutionizing how developers build applications powered by large language models. Created by Harrison Chase in 2022, LangChain provides modular components that help connect AI models to external data sources, tools, and reasoning capabilities. We break down LangChain's key components—models, prompts, memory, chains, agents, and tools—explaining how they work together to overcome the inherent limitations of language models. The discussion highlights popular applications, including retrieval-augmented generation systems that connect AI to private data, conversational agents that can use external tools, document analysis solutions, and personal assistants. While emphasizing LangChain's flexibility and growing ecosystem, we also address its limitations and alternatives like Haystack, LlamaIndex, and Semantic Kernel. Whether you're a developer looking to build AI applications or simply interested in how modern AI systems work, this episode provides a comprehensive introduction to one of the most important frameworks in today's AI development landscape.

    Find more information at https://www.ibm.com/think/topics/langchain Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Amanda Downie
    7 min
  • What is LangGraph?
    This episode of Techsplainers explores LangGraph, an open-source framework that extends LangChain to enable stateful, cyclic workflows for large language model applications. Built by the team behind LangChain, LangGraph provides developers with the tools to create AI agents capable of complex reasoning and multi-step processes that can adapt based on intermediate results. We break down the key components of LangGraph—nodes, edges, state management, conditional logic, and cycle detection—explaining how they work together to enable more human-like problem-solving approaches. The discussion highlights practical applications including the ReAct pattern for combining reasoning with action-taking, multi-agent systems where specialized AI agents collaborate, iterative refinement workflows for creative tasks, and tree-of-thought reasoning for complex problem solving. While acknowledging the challenges of designing and debugging graph-based workflows, the episode emphasizes how LangGraph's ability to maintain state throughout complex processes opens up new possibilities for sophisticated AI applications that can plan, backtrack, and adapt to changing conditions.

    Find more information at https://www.ibm.com/think/topics/langgraph Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Amanda Downie
    6 min
  • What is LangFlow?
    This episode of Techsplainers explores Langflow, an innovative open-source visual programming tool that makes building AI applications with large language models more accessible and intuitive. Unlike traditional coding approaches, Langflow provides a drag-and-drop interface for designing LangChain workflows, allowing both developers and non-technical users to create sophisticated AI applications. We examine Langflow's key components—its comprehensive component library, interactive visual canvas, real-time testing capabilities, Python code export functionality, and built-in debugging tools. The discussion highlights how Langflow democratizes AI development through its visual approach, enabling rapid prototyping, educational opportunities, cross-functional collaboration, and easy experimentation with AI concepts. While acknowledging limitations for extremely complex applications and the continued importance of understanding fundamental concepts, the episode emphasizes Langflow's value in lowering barriers to entry for AI application development and bridging the gap between technical and business teams.

    Find more information at https://www.ibm.com/think/topics/langflow Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Amanda Downie
    8 min
  • What is MetaGPT?
    This episode of Techsplainers explores MetaGPT, an innovative open-source framework that orchestrates multiple AI agents to function as a virtual software development team. Created by DeepWisdom AI, MetaGPT transforms simple requirements into working applications by simulating specialized roles like product managers, architects, programmers, and testers—each played by large language models configured for specific tasks. We examine how MetaGPT's Standard Operating Procedures guide these agents through the entire software development lifecycle, from requirement analysis and system design to implementation and testing. The discussion highlights the framework's structured approach to collaborative AI, including its multi-agent system, communication protocols, and execution environment. While acknowledging current limitations in handling complex systems and the continued need for human oversight, the episode emphasizes MetaGPT's significance in advancing AI-powered software development and providing a glimpse into how autonomous agent systems might transform creative and technical workflows in the future.

    Find more information at https://www.ibm.com/think/topics/metagpt Find more episodes https://www.ibm.biz/techsplainers-podcast

    Narrated by Amanda Downie
    7 min

About Techsplainers by IBM

From the publisher's feed

Introducing the Techsplainers by IBM podcast, your new podcast for quick, powerful takes on today’s most important AI and tech topics. Each episode brings you bite-sized learning designed to fit your day, whether you’re driving, exercising, or just curious for something new.