Educational Offerings - LangChain Education provides various learning methods, including courses, YouTube videos, and documentation [1] - LangChain Academy offers three types of courses: Foundational, Project, and Quickstart [1] Course Types - Foundational courses offer methodical learning from introduction to mastery and require more time to complete [2] - Project courses guide users through building specific projects, such as a Deep Research agent, and can typically be completed in a few hours [2] - Quickstart courses provide a quick introduction or review of a topic [2] Additional Resources - LangChain publishes educational videos on YouTube covering current topics, product features, and in-depth series [3] - LangChain provides extensive documentation with examples and step-by-step instructions [3]

Open Source LLM Framework - A framework connects any LLM to any MCP server (open-source) [1] - The framework enables building custom MCP Agents without closed-source apps [1] - Compatible with Ollama, LangChain, etc [1] - Allows building 100% local MCP clients [1]

LLM & MCP Integration - A framework enables connecting any LLM to any MCP server [1] - The framework facilitates building custom MCP Agents without relying on closed-source applications [1] - It is compatible with tools like Ollama and LangChain [1] - The framework allows building 100% local MCP clients [1]

RT Avi Chawla (@_avichawla)After MCP, A2A, & AG-UI, there's another Agent protocol (open-source).ACP (Agent Communication Protocol) is a standardized, RESTful interface for Agents to discover and coordinate with other Agents, regardless of their framework (CrewAI, LangChain, etc.).Here's how it works:- Build your Agents and host them on ACP servers.- The ACP server will receive requests from the ACP Client and forward them to the Agent.- ACP Client itself can be an Agent to intelligently route requests to t ...

Agent Communication Protocols - ACP (Agent Communication Protocol) is presented as a new open-source protocol for agent communication, offering a standardized RESTful interface [1] - The protocol facilitates agent discovery and coordination across different frameworks like CrewAI and LangChain [1] - An ACP Client can function as an intelligent router for requests to agents, similar to MCP Client [1] ACP vs A2A - ACP is designed for local-first, low-latency communication, while A2A is optimized for web-native, cross-vendor interoperability [3] - ACP utilizes a RESTful interface for easier integration, whereas A2A supports more flexible interactions [3] - ACP is suited for controlled, edge, or team-specific environments, while A2A excels in broader cloud-based collaboration [3] Development and Deployment - The industry is encouraged to build ACP-compliant agents using frameworks like CrewAI and Smolagents [3] - Agents can be chained in sequential and hierarchical workflows, managed by ACP clients [3] - Agents can be imported to a public registry for easy discovery [3] - DeeplearningAI offers a course on building and hosting agents on ACP servers [2]

Core Insights - The article discusses the challenges faced by Go developers in creating AI applications, highlighting the need for a native AI development experience tailored for Go language [1][2] - It introduces a new framework, Eino, aimed at enabling Go developers to build AI agents and applications more efficiently [4][5] Group 1: Event Overview - A live demonstration will be organized for backend developers, focusing on the Deep Research application, Deerflow, which utilizes LangChain and LangGraph [4] - The goal of the live session is to build a complete AI application from scratch using the Eino framework, showcasing the architecture and design principles [4][5] Group 2: Expert Involvement - Two engineers from ByteDance will participate in the event, with one acting as the "architect decoder" to explain the design of Deerflow, and the other as the "Go AI application master" to demonstrate the implementation using Eino [5] - This collaboration aims to provide insights into defining powerful AI agents and the practical application of the Eino framework [5][7] Group 3: Target Audience - The event is targeted at Go developers looking to enhance their competitive edge in AI, AI/LLM application developers seeking efficient frameworks, and backend engineers curious about AI technology [7] - Participants are encouraged to engage with the content if they are passionate about creating intelligent solutions through code [7] Group 4: Event Details - The live session is scheduled for July 9, 2025, at 7:30 PM, with opportunities for participants to win custom prizes [8] - Registration is available through a QR code for reminders and exclusive materials [8] Group 5: Conclusion - The article emphasizes the potential for a significant shift in the Go language's capabilities in the AI agent domain, promising an exciting event for attendees [9]

Core Insights - The article discusses the transformation in the AI industry, emphasizing the shift from individual AI model usage to a collaborative network of agents, termed as "Agent collaboration network" [8][10][27] - It highlights the urgency for AI professionals to adapt their skills from prompt engineering to organizing and managing AI collaborations, as traditional skills may become obsolete [9][21][30] Group 1: Industry Trends - The AI landscape is evolving towards a multi-agent system where agents communicate and collaborate autonomously, moving away from reliance on human prompts [27][14] - The emergence of protocols like MCP (Multi-agent Communication Protocol) and A2A (Agent-to-Agent) is facilitating this transition, allowing for standardized communication between different AI systems [36][37] - Major companies like Alibaba, Tencent, and ByteDance are rapidly developing platforms that support these new protocols, enabling easier integration and deployment of AI agents [38][39] Group 2: Skills Transformation - AI professionals need to transition from being prompt engineers to "intent architects," focusing on defining task languages and collaboration protocols for agents [29][30] - The role of AI practitioners is shifting from using agents to organizing and managing multiple agents, requiring a new mindset akin to building a digital team [30][31] - There is a call for professionals to learn about agent frameworks, communication protocols, and how to register their tools as agent capabilities within larger networks [33][34] Group 3: Practical Applications - Various platforms and frameworks are emerging that allow AI professionals to practice and implement these new skills, such as LangGraph, AutoGen, and CrewAI [41] - The article emphasizes that the infrastructure for agent protocols is being established, providing opportunities for AI professionals to engage with these technologies [41][42] - The ongoing development of these systems is likened to the early days of TCP/IP, suggesting that those who adapt early will have a competitive advantage in the evolving AI landscape [42]

曲凯： 今天我们请来了国内强化学习 (RL) 领域的专家吴翼，吴翼目前是清华大学交叉信息研究院助理教授，他曾经在 OpenAI 工作过，算是国内最早研究强化学 习的人之一，我们今天就争取一起把 RL 这个话题给大家聊透。 首先吴翼能不能简单解释一下，到底什么是 RL？ 因此，RL 其实更通用一些，它的逻辑和我们在真实生活中解决问题的逻辑非常接近。比如我要去美国出差，只要最后能顺利往返，中间怎么去机场、选什么航 司、具体坐哪个航班都是开放的。 但 RL 很不一样。 RL 最早是用来打游戏的，而游戏的特点和分类问题有两大区别。 第一，游戏过程中有非常多的动作和决策。比如我们玩一个打乒乓球的游戏，发球、接球、回球，每一个动作都是非标的，而且不同的选择会直接影响最终的结 果。 第二，赢得一场游戏的方式可能有上万种，并没有唯一的标准答案。 所以 RL 是一套用于解决多步决策问题的算法框架。它要解决的问题没有标准答案，每一步的具体决策也不受约束，但当完成所有决策后，会有一个反馈机制来评 判它最终做得好还是不好。 吴翼： RL 是机器学习这个大概念下一类比较特殊的问题。 传统机器学习的本质是记住大量标注过正确答案的数据对。 ...

Core Viewpoint - The article discusses the emerging interoperability standards in the AI field, particularly focusing on the Model Context Protocol (MCP) and its significance in enhancing AI model connectivity and collaboration [3][6][9]. Group 1: MCP Overview - MCP was proposed by Anthropic as an open standard to enable seamless interaction between large language models and various external data sources and tools, akin to a "universal connector" in the AI realm [7][9]. - The support from major players like OpenAI and Google has accelerated MCP's transition from a potential proposal to a widely accepted standard, marking a significant step towards unified and efficient AI application development [7][9][16]. Group 2: Benefits of MCP - MCP's core value lies in its standardization, which allows any AI model to interact with external resources through lightweight MCP servers, addressing the fragmentation issue of custom integration for each model and tool [9][10]. - The protocol enhances the AI's ability to perform complex tasks by enabling it to combine multiple MCP servers, thus facilitating cross-service collaboration and more sophisticated agent behaviors [10][11]. Group 3: Real-World Applications - MCP allows AI to directly query databases and interact with productivity tools, significantly improving workflow integration [10][11]. - Developers like Codeium have integrated MCP into their tools, enabling AI to perform a variety of development tasks beyond simple code completion, thus enhancing the capabilities of integrated development environments (IDEs) [12][11]. Group 4: Limitations and Future Potential - While MCP shows promise, it currently has limitations in specialized fields, where AI may struggle with complex tasks requiring precise understanding, indicating that its effectiveness is contingent on the capabilities of the MCP servers and the AI models [15][16]. - The article suggests that as more industry players adopt MCP and build related ecosystems, it could become a foundational technology for integrating AI with existing software and services, similar to how browser extension APIs transformed web interactions [16].