Войти
  • 17300Просмотров
  • 5 месяцев назадОпубликованоGoogle Cloud Tech

Agents, their frameworks and when to use them | The Agent Factory Podcast

Start building with Agent Development Kit today! → Welcome to the very first episode of The Agent Factory, the podcast that goes beyond the hype and dives into building production-ready AI agents. Join hosts Shir Meir Lador and Ivan Nardini as they help you navigate the world of agentic AI. By the end of this episode, you'll have a solid understanding of what a true AI agent is, why you need a framework to build one, and the key trade-offs between popular tools like LangChain, LangGraph, crewAI, and the Agent Development Kit (ADK). This will equip you to choose the right tools for your next project. In This Episode, You'll Learn: Agent Industry Pulse: We cover the latest updates in the agent world, including new releases from LangGraph that improve multi-agent performance , crewAI's focus on asynchronous tasks , LlamaIndex's new Spreadsheet Agent , Databricks' Agent Bricks , new code assistants from Anthropic and Mistral, and the v1 production-ready release of Google's Agent Development Kit (ADK). The Factory Floor: We provide a foundational deep dive into what an AI agent is by comparing a simple FAQ chatbot to a dynamic, reasoning Customer Support Agent that can form and adapt its own plans. We also explore why frameworks are critical for managing state and providing observability and compare between the different frameworks for agent development Langraph, Langchain and CrewAI. Special Guest : Julia Wiesinger: Julia, from the Google Agent Development Kit team, joins us to discuss where the ADK fits into the ecosystem. She explains how it provides an "opinionated assembly line" for production agents by simplifying integration, offering a prebuilt Web UI, and enabling multimodal interactions with its Live API. Developer Q&A: We tackle your questions about agent architecture, including when to use a Sub-Agent vs. an Agent-as-a-Tool , the trade-offs between deploying on Cloud Run vs. GKE vs. Vertex AI Agent engine, and the multifaceted approach required to properly evaluate an agent's performance beyond its final response. About The Agent Factory: "The Agent Factory" is a video-first technical podcast for developers, by developers, focused on building production-ready AI agents. We'll explore how to design, build, deploy, and manage agents that bring real value using the latest open-source tools and frameworks. Going beyond basic AI/ML concepts, the podcast will delve into the practical aspects of agentic AI, covering everything from single-purpose agents for task automation to complex multi-agent systems. Connect with Us & Ask Questions: Leave your questions and feedback in the YouTube comments below! Join the conversation on social media with the hashtag #TheAgentFactory. Connect with the community at the Google Cloud Community forum → Resources: LangGraph Release Week Recap → Benchmarking Multi-Agent Architectures → Release→ Agent Development Kit (ADK) Blog → Agent Development Kit (ADK)→ LangChain→ LangGraph → crewAI→ Evaluation white paper→ Subscribe to Google Cloud Tech → #AIAgents #GoogleCloud #ADK Speakers: Shir Meir Lador, Ivan Nardini Products Mentioned: Vertex AI, Agent Developer Kit (ADK), Cloud Run, Google Kubernetes Engine (GKE)