AI Integration in Communication

LangMem is a new open-source library that gives LLM agents long-term memory and it’s refreshingly easy to use. It’s built for developers working with LangGraph or custom agents, and it solves a persistent problem: how to make agents remember and adapt across sessions without bloated prompts or manual hacks. LangMem introduces a clean memory API that works with any storage backend and includes tools for: (1) Storing important information during conversations—agents decide what matters and when to save it (2) Searching memory when relevant—retrieving facts, preferences, or prior context (3) Running background memory consolidation—automatically refining and updating knowledge over time It integrates natively with LangGraph’s memory store, but you can also plug it into your own stack using Postgres, Redis, or in-memory stores. This design is especially useful for building agents that need to: -> Personalize interactions across sessions -> Maintain consistency in long-running workflows -> Adapt behavior based on evolving user input Unlike Mem0, which requires explicit memory updates, LangMem handles memory automatically in the background, storing and retrieving key details as needed, and integrates with LangGraph out of the box. GitHub repo https://lnkd.in/gj6i3Q8p This repo and 40+ curated open-source frameworks and libraries for AI agents builders in my recent post https://lnkd.in/g3fntJVc

OpenAI is hiring for storytelling and communications roles, and paying almost $400K a year. The irony is perfect. Everyone uses ChatGPT to write the exact type of content OpenAI now needs humans to create. Draft emails, social media posts, marketing copy, all the stuff they are posting job descriptions for. Two things are happening at the same time. The value of authentic storytelling keeps rising while AI makes generic content, well, more generic.   Companies that can tell compelling stories about their products, vision, and impact will win. The platforms do not matter - TikTok, Instagram, LinkedIn, your own site. What matters is the ability to connect with people through narrative. But here is the catch. AI can generate infinite amounts of content, but cannot create a genuine connection. It can mimic tone and style, but cannot share real experiences or build an authentic community. OpenAI hiring human storytellers while their product replaces human writers reveals the paradox. The better AI gets at producing content, the more valuable the human perspective becomes. The companies winning at lead generation and community building are not using AI to replace their communications, but to handle the boring stuff so humans can focus on the stories that actually matter. This creates a weird dynamic. AI democratizes basic content creation but increases the premium for authentic voice and genuine insight. The jobs getting posted are not for people who can write. They are for people who can think, connect, and communicate ideas that matter.

𝗔𝟮𝗔 (𝗔𝗴𝗲𝗻𝘁-𝘁𝗼-𝗔𝗴𝗲𝗻𝘁) 𝗮𝗻𝗱 𝗠𝗖𝗣 (𝗠𝗼𝗱𝗲𝗹 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹) are two emerging protocols designed to facilitate advanced AI agent systems, but they serve distinct roles and are often used together in modern agentic architectures. 𝗛𝗼𝘄 𝗧𝗵𝗲𝘆 𝗪𝗼𝗿𝗸 𝗧𝗼𝗴𝗲𝘁𝗵𝗲𝗿 Rather than being competitors, 𝗔𝟮𝗔 𝗮𝗻𝗱 𝗠𝗖𝗣 𝗮𝗿𝗲 𝗰𝗼𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗮𝗿𝘆 𝗽𝗿𝗼𝘁𝗼𝗰𝗼𝗹𝘀 that address different layers of the agent ecosystem: • 𝗔𝟮𝗔 is about agents collaborating, delegating tasks, and sharing results across a distributed network. For example, an orchestrating agent might delegate subtasks to specialized agents (analytics, HR, finance) via A2A25. • 𝗠𝗖𝗣 is about giving an agent (often an LLM) structured access to external tools and data. Within an agent, MCP is used to invoke functions, fetch documents, or perform computations as needed.    𝗧𝘆𝗽𝗶𝗰𝗮𝗹 𝗪𝗼𝗿𝗸𝗳𝗹𝗼𝘄 𝗘𝘅𝗮𝗺𝗽𝗹𝗲: • A user submits a complex request. • The orchestrating agent uses 𝗔𝟮𝗔 to delegate subtasks to other agents. • One of those agents uses 𝗠𝗖𝗣 internally to access tools or data. • Results are returned via A2A, enabling end-to-end collaboration25.    𝗗𝗶𝘀𝘁𝗶𝗻𝗰𝘁 𝗦𝘁𝗿𝗲𝗻𝗴𝘁𝗵𝘀 • 𝗔𝟮𝗔 𝗲𝘅𝗰𝗲𝗹𝘀 𝗮𝘁:   Multi-agent collaboration and orchestration   Handling complex, multi-domain workflows   Allowing independent scaling and updating of agents   Supporting long-running, asynchronous tasks54 • 𝗠𝗖𝗣 𝗲𝘅𝗰𝗲𝗹𝘀 𝗮𝘁:   Structured tool and data integration for LLMs   Standardizing access to diverse resources   Transparent, auditable execution steps   Single-agent scenarios needing a precise tool    𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗮𝗹 𝗔𝗻𝗮𝗹𝗼𝗴𝘆 • 𝗠𝗖𝗣 is like a 𝘶𝘯𝘪𝘷𝘦𝘳𝘴𝘢𝘭 𝘤𝘰𝘯𝘯𝘦𝘤𝘵𝘰𝘳 (USB-C port) between an agent and its tools/data. • 𝗔𝟮𝗔 is like a 𝘯𝘦𝘵𝘸𝘰𝘳𝘬 𝘤𝘢𝘣𝘭𝘦 connecting multiple agents, enabling them to form a collaborative team.    𝗦𝗲𝗰𝘂𝗿𝗶𝘁𝘆 𝗮𝗻𝗱 𝗖𝗼𝗺𝗽𝗹𝗲𝘅𝗶𝘁𝘆 𝗖𝗼𝗻𝘀𝗶𝗱𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝘀 • 𝗔𝟮𝗔 introduces many endpoints and requires robust authentication and authorization (OAuth2.0, API keys). • 𝗠𝗖𝗣 needs careful sandboxing of tool calls to prevent prompt injection or tool poisoning. Both are built with enterprise security in mind.    𝗜𝗻𝗱𝘂𝘀𝘁𝗿𝘆 𝗔𝗱𝗼𝗽𝘁𝗶𝗼𝗻 • 𝗔𝟮𝗔: Google, Salesforce, SAP, LangChain, Atlassian, Cohere, and others are building A2A-enabled agents. • 𝗠𝗖𝗣: Anthropic (Claude Desktop), Zed, Cursor AI, and tool-based LLM UIs.   Modern agentic systems often combine both: 𝗔𝟮𝗔 𝗳𝗼𝗿 𝗶𝗻𝘁𝗲𝗿-𝗮𝗴𝗲𝗻𝘁 𝗼𝗿𝗰𝗵𝗲𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻, 𝗠𝗖𝗣 𝗳𝗼𝗿 𝗶𝗻𝘁𝗿𝗮-𝗮𝗴𝗲𝗻𝘁 𝘁𝗼𝗼𝗹 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻. This layered approach supports scalable, composable, and secure AI applications.

You can now connect your AI agent to 250+ tools using MCP 🤯 Without writing a single line of code. Composio just introduced fully managed MCP servers with built-in auth. Most teams building AI agents face the same problem: ↳ Setting up reliable MCP servers is hard  ↳ Authentication flows are complex  ↳ Server maintenance is a headache  ↳ Each integration requires custom work This is why Composio's MCP servers makes so much sense. They've built: ↳ Fully managed MCP servers for tools like Slack, Notion, and Linear ↳ Seamless auth handling (OAuth, API keys, JWT) ↳ 20,000+ pre-built API actions ↳ Few-clicks connections to Claude, Cursor, Windsurf, and AI agents I watched their demo - it's impressively simple. For Cursor: Search your app, copy a URL, paste it into settings. For Claude Desktop: One terminal command connects Gmail The real power is what happens next. Your AI agent can now: → Send emails through Gmail → Create tasks in Linear → Search documents in Notion → Post messages in Slack → Update records in Salesforce All while you chat naturally with it. Think about what this means for productivity. Tasks that used to require context switching between 5+ apps  Can now happen in a single conversation with your agent. No more building custom integrations.  No more authentication headaches.  No more server maintenance. The teams moving fastest right now are the ones  Leveraging AI agents connected to their work tools. Are you still building integrations from scratch? Or are you ready to plug into a solution that just works? Get ahead with Composio's pre-built MCP integrations: https://mcp.composio.dev/ P.S. I create these AI tutorials and opensource them for free. Your 👍 like and ♻️ repost helps keep me going. Don't forget to follow me Shubham Saboo for daily tips and tutorials on LLMs, RAG and AI Agents.

Google just launched Agent2Agent (A2A) protocol that could quietly reshape how AI systems work together. If you’ve been watching the agent space, you know we’re headed toward a future where agents don’t just respond to prompts. They talk to each other, coordinate, and get things done across platforms. Until now, that kind of multi-agent collaboration has been messy, custom, and hard to scale. A2A is Google’s attempt to fix that. It’s an open standard for letting AI agents communicate across tools, companies, and systems, that securely, asynchronously, and with real-world use cases in mind. What I like about it: - It’s designed for agent-native workflows (no shared memory or tight coupling) - It builds on standards devs already know: HTTP, SSE, JSON-RPC - It supports long-running tasks and real-time updates - Security is baked in from the start - It works across modalities- text, audio, even video But here’s what’s important to understand: A2A is not the same as MCP (Model Context Protocol). They solve different problems. - MCP is about giving a single model everything it needs- context, tools, memory, to do its job well. - A2A is about multiple agents working together. It’s the messaging layer that lets them collaborate, delegate, and orchestrate. Think of MCP as helping one smart model think clearly. A2A helps a team of agents work together, without chaos. Now, A2A is ambitious. It’s not lightweight, and I don’t expect startups to adopt it overnight. This feels built with large enterprise systems in mind, teams building internal networks of agents that need to collaborate securely and reliably. But that’s exactly why it matters. If agents are going to move beyond “cool demo” territory, they need real infrastructure. Protocols like this aren’t flashy, but they’re what make the next era of AI possible. The TL;DR: We’re heading into an agent-first world, and that world needs better pipes. A2A is one of the first serious attempts to build them. Excited to see how this evolves.

🤖 The Leadership Charade: When AI Makes You Sound Authentic (But You're Not) Imagine streamlining annual reviews with cutting-edge AI. Sounds revolutionary, right? But at what cost to employee connection and authenticity? "I automated my team's performance reviews using AI to save time." An executive client shared this with me recently, beaming with pride at their efficiency. Their smile vanished when I asked: "But do your people feel truly seen in those reviews?" We're crossing into dangerous territory: using technology to sound authentic while skipping the human work of truly connecting with our people. This isn't about rejecting AI. It's about being intentional about what only humans can do well. 🌟 AI can craft your communications, analyze your metrics, and optimize your schedule. But it absolutely cannot: • Build genuine psychological safety • Practice radical kindness when someone's struggling • Notice the unspoken dynamics in your team • Embody the vulnerable leadership that fosters belonging The most alarming pattern I've witnessed in my executive coaching? Leaders using AI to manufacture what research calls "toxic positivity" defined as perfectly crafted, upbeat messages that lack the messy authenticity of real human connection. When our leadership becomes too polished, too perfect, we lose the beautiful imperfection that makes us human. We lose trust. Here's the paradox: As AI makes communication more efficient, authentic connection becomes more valuable. 💥 The radically kind approach: ⚡ Draft important messages yourself first - even when imperfect ⚡ Ask: "Would my team recognize my voice in this?" ⚡ Include observations that only you would notice ⚡ Share a genuine challenge you're navigating Technology is a magnificent tool. But tools should amplify our humanity, not replace it. In an AI world, your humanity isn't a leadership weakness. It's your superpower. ⚡ And, Yes!, I use AI tools every day of my work and personal life. I continue to learn every day and stay Constantly Curious. Let's learn from each other authentically. Let's Celebrate being Human as much as we celebrate new Tech

I'm knee deep this week putting the finishing touches on my new Udemy course on "AI for People Managers: Lead with confidence in an AI-enabled workplace". After working with hundreds of managers cautiously navigating AI integration, here's what I've learned: the future belongs to leaders who can thoughtfully blend AI capabilities with genuine human wisdom, connection, and compassion. Your people don't need you to be the AI expert in the room; they need you to be authentic, caring, and completely committed to their success. No technology can replicate that. And no technology SHOULD. The managers who are absolutely thriving aren't necessarily the most tech-savvy ones. They're the leaders who understand how to use AI strategically to amplify their existing strengths while keeping clear boundaries around what must stay authentically human: building trust, navigating emotions, making tough ethical calls, having meaningful conversations, and inspiring people to bring their best work. Here's the most important takeaway: as AI handles more routine tasks, your human leadership skills become MORE valuable, not less. The economic value of emotional intelligence, empathy, and relationship building skyrockets when machines take over the mundane stuff. Here are 7 principles for leading humans in an AI-enabled world: 1. Use AI to create more space for real human connection, not to avoid it 2. Don't let AI handle sensitive emotions, ethical decisions, or trust-building moments 3. Be transparent about your AI experiments while emphasizing that human judgment (that's you, my friend) drives your decisions 4. Help your people develop uniquely human skills that complement rather than compete with technology. (Let me know how I can help. This is my jam.) 5. Own your strategic decisions completely. Don't hide behind AI recommendations when things get tough 6. Build psychological safety so people feel supported through technological change, not threatened by it 7. Remember your core job hasn't changed. You're still in charge of helping people do their best work and grow in their careers AI is just a powerful new tool to help you do that job better, and to help your people do theirs better. Make sure it's the REAL you showing up as the leader you are. #AI #coaching #managers

AI is reshaping how we will work but most people are still stuck thinking "chatbot." I sat down with Aakash Gupta to break down AI agents for you. This was my first long-format podcast. I have a long way to go and a lot to improve. I still panic in front of cameras 🤪 but I genuinely enjoyed this first one. We covered: 1/ Why AI agents will transform every role 2/ How enterprises are really using AI (hint: it's not what you think) 3/ The career moves that matter in the AI era Here’s what we cover: 1. Why AI agents will transform every role: ▪️ We’re moving from the “chatbot era” to true automation. Agents don’t just respond; they think, plan, act, and reflect to complete enterprise workflows. ▪️ Most professionals will manage 10–20 AI agents within 3–5 years. Each specialized for things like competitive research, user feedback synthesis, or prototyping. ▪️ Start with no-code tools like Langflow, Lindy or N8N. Graduate to frameworks like LangGraph, Agno or CrewAI when you hit enterprise-level complexity. 2. How enterprises are really using AI: ▪️ 90% of use cases in the first post-ChatGPT year were RAG systems. Connecting AI to internal data; not just general knowledge tasks. ▪️ It’s not about saving costs. It’s about doing exponentially more. Companies that embrace this will outpace the rest. ▪️ Vision RAG is the a big unlock: interpreting charts, diagrams, and visual data that text-only systems miss. 3. The career moves that matter in the AI era: ▪️ Technical literacy is now table stakes. You won’t know what’s possible with AI unless you build something. ▪️ Prototype-first beats slide decks. Aakash shared how a working demo landed him a game-changing project. ▪️ Go deep where AI meets your domain. Specific > generic. Also in there: the story of the meeting that changed my career and how I moved from Spain to Silicon Valley. I hope you listen and enjoy the episode. Send me questions for any topic that was not clear and feedback to do better next time! 🎬 Watch on Youtube: https://lnkd.in/gtRX8DQ6 🎧 Listen: - Spotify: https://lnkd.in/grfTHWf5 - Apple: https://lnkd.in/gRddeRdy

Some brands are slapping AI on their marketing like a shiny sticker — it looks modern, but it doesn’t mean anything. Seeing so many brands use AI in their taglines at a conference I attended recently and then seeing Dove publish AI role with real beauty got me thinking about how as brand leaders, we need to frame the role of AI into our brand strategy. If you’re a CMO or brand leader, here’s the question to ask: Can AI help us deliver our brand promise better than anyone else — and make that difference obvious to customers? If the answer’s yes, you’ve got something worth building. If the answer’s no, you’re just playing with toys and you have more work to do with your team. Here’s a framework I’ve been playing with to make AI work harder — not just for productivity, but for true brand differentiation, ie. adding value. As a placeholder, I am calling it the A.I.D.E.A. Framework: A — Anchor in Your Brand Promise Start with what you stand for — your why. AI should enhance your ability to deliver that promise, not distract from it. Ex: Dove used AI to uphold its Real Beauty values, creating standards to fight unrealistic beauty filters. I posted about this yesterday. I — Identify Distinctive Touchpoints Pinpoint the moments where your brand naturally stands apart in the customer journey. Then ask: where could AI enhance that difference? Ex: Pedigree used AI to turn everyday ads into hyper-local dog adoption campaigns. D — Design On-Brand Experiences Your AI outputs (interfaces, language, tone, visuals) should feel unmistakably like you. AI can scale your brand voice — if you train it right. Ex: L’Oréal’s beauty assistant reflects their expertise and inclusivity, not just product recs. E — Execute Transparently and Ethically Build trust into your AI strategy. Be clear with consumers when and how AI is used — and why it benefits them. Ex: Salesforce emphasized data security and transparency as core features of Einstein GPT. A — Amplify with Storytelling Showcase how AI deepens your promise. Don’t just say “we use AI” — say what it lets you do for people that no one else can. Ex: Coca-Cola’s “Real Magic” AI campaign let fans co-create with Coke — making creativity part of the brand. TL;DR for Brand Leaders: AI won’t make you different, at least not yet. But if you’re already different — it can make you unmistakable. Would you use this with your team? #AIinMarketing #BrandStrategy #CMO #GenerativeAI #BrandDifferentiation #MarketingLeadership Elizabeth Oates Priti Mehra Raul Ruiz David Bernardino Lauren Morgenstein Schiavone Kristi Zuhlke

How do we make AI agents truly useful in the enterprise? Right now, most AI agents work in silos. They might summarize a document, answer a question, or write a draft—but they don’t talk to other agents. And they definitely don’t coordinate across systems the way humans do. That’s why the A2A (Agent2Agent) protocol is such a big step forward. It creates a common language for agents to communicate with each other. It’s an open standard that enables agents—whether they’re powered by Gemini, GPT, Claude, or LLaMA—to send structured messages, share updates, and work together. For enterprises, this solves a very real problem: how do you connect agents to your existing workflows, applications, and teams without building brittle point-to-point integrations? With A2A, agents can trigger events, route messages through a shared topic, and fan out information to multiple destinations—whether it’s your CRM, data warehouse, observability platform, or internal apps. It also supports security, authentication, and traceability from the start. This opens up new possibilities: An operations agent can pass insights to a finance agent A marketing agent can react to real-time product feedback A customer support agent can pull data from multiple systems in one seamless thread I’ve been following this space closely, and I put together a visual to show how this all fits together—from local agents and frameworks like LangGraph and CrewAI to APIs and enterprise platforms. The future of AI in the enterprise won’t be driven by one single model or platform—it’ll be driven by how well these agents can communicate and collaborate. A2A isn’t just a protocol—it’s infrastructure for the next generation of AI-native systems. Are you thinking about agent communication yet?