Armand Ruiz's Linkedin Analytics

Yeah! It's official: I'll be at NVIDIA GTC this year; this is like the Super Bowl of AI! Catch me in San Jose from March 17–21, 2025, I hope to see you there! With over 1,000 sessions and a ton of epic announcements, it’s gonna be great. Scope out the agenda here: https://lnkd.in/gatvhSqP. Can’t make it? No worries — tune into the main keynote on YouTube and jump into the livestream for all the action

Llama Stack: Go from Zero to Hero Meta has published a great guide that takes you from zero to production with Llama Stack. Whether you're a beginner or looking to level up your AI development skills, this guide is a great roadmap. What makes this guide special? - Detailed walkthrough from setup to deployment - Step-by-step instructions for local AI model configuration - Learn to use Ollama for local inference - Practical examples of model interaction Multiple API guides covering: - Text Generation - Prompt Engineering - Tool Calling - Image Interaction - Safety Protocols The guide breaks down complex concepts into digestible steps, helping you: ✅ Set up your development environment ✅ Configure Llama Stack ✅ Run your first AI model ✅ Explore advanced API capabilities Pro developers will appreciate the flexibility of prototyping locally and then seamlessly deploying to the cloud. Start Now! Here's the link: https://lnkd.in/g_DSGWyV

Check out this repo with 100s of MCP Servers, including reference implementations and community-built solutions. MCP enables secure, controlled access for LLMs to tools and data sources. Each of them uses either the TypeScript MCP SDK or Python MCP SDK, providing a solid starting point for developers. Here are some of my favorite ones: 𝟭/ 𝗔𝗪𝗦 𝗞𝗕 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 – Secure enterprise knowledge retrieval from AWS Knowledge Base. 𝟮/ 𝗚𝗼𝗼𝗴𝗹𝗲 𝗗𝗿𝗶𝘃𝗲 – File access and search capabilities for enterprise collaboration. 𝟯/ 𝗣𝗼𝘀𝘁𝗴𝗿𝗲𝗦𝗤𝗟 – Read-only database access with schema inspection for structured data queries. 𝟰/ 𝗚𝗿𝗮𝗳𝗮𝗻𝗮 – Investigate incidents and query dashboards for system monitoring. 𝟱/ 𝗦𝗹𝗮𝗰𝗸 – Enterprise messaging and channel management for team communication. 𝟲/ 𝗦𝘁𝗿𝗶𝗽𝗲 – Financial transactions, payments, and API integration for businesses. 𝟳/ 𝗖𝗹𝗼𝘂𝗱𝗳𝗹𝗮𝗿𝗲 – Manage and configure Cloudflare resources for security and performance. 𝟴/ 𝗦𝗮𝗹𝗲𝘀𝗳𝗼𝗿𝗰𝗲 𝗠𝗖𝗣 – CRM data and metadata interaction for sales and customer management. 𝟵/ 𝗡𝗲𝗼𝟰𝗷 – Graph database server for complex enterprise data relationships. 𝟭𝟬/ 𝗦𝗻𝗼𝘄𝗳𝗹𝗮𝗸𝗲 – Secure access and interaction with cloud-based data warehouses. Link to the MCP serve catalog: https://lnkd.in/gmYCvhST Happy building! 🤖

Out of the box, AI agents fail ~40% of the time in unexpected ways when performing enterprise tasks. Initial implementations often struggle with unpredictable scenarios and lack robust intelligence. I'm talking about AI agents that can navigate across multiple enterprise apps and handle tons of unstructured data, not just solving specific issues like parsing tables in documents but effectively managing complex, unstructured situations. That's the real challenge with sophisticated AI agents today. 👉 𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗮𝗹𝗹 𝗼𝗳 𝘆𝗼𝘂: How are you addressing this issue currently? What approaches have you found effective?

welcome to the future

NVIDIA’s GTC week is here! This is the tech event of the year. I remember watching, as a teenager, the excitement Apple events used to generate in the early 2000s. Now, that excitement is all about AI—and it’s happening at GTC. If you’re not tuned in, you’re missing out on the pulse of technological progress! Follow me at Armand Ruiz to stay updated on the highlights and key takeaways.

Meet NVIDIA AgentIQ, an open-source toolkit designed to streamline the development and enhance the performance of complex AI agent systems. This wasn't even announced in the main keynote yesterday, and it is such an exciting project. This is how it works: 𝟭/ 𝗦𝗶𝗺𝗽𝗹𝗶𝗳𝘆 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁: Quickly prototype agentic AI applications with our configuration builder and flexible agent frameworks. 𝟮/ 𝗔𝗰𝗰𝗲𝘀𝘀 𝗣𝗼𝘄𝗲𝗿𝗳𝘂𝗹 𝗧𝗼𝗼𝗹𝘀: Utilize NVIDIA AI Blueprints and advanced RAG architectures to build scalable, accurate AI agents. 𝟯/ 𝗔𝗰𝗰𝗲𝗹𝗲𝗿𝗮𝘁𝗲 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗲𝘀: Fine-grained telemetry optimizes performance, enhancing efficiency without infrastructure upgrades. 𝟰/ 𝗜𝗻𝗰𝗿𝗲𝗮𝘀𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆: Evaluate and optimize agentic systems with comprehensive metrics and integration capabilities. You can learn more and access the documentation here: https://lnkd.in/ei66TRPt

Not enough people are talking about this: SLMs are giving our customers better AI performance at 98% lower cost This is even more critical in agentic use cases, as they require more compute.

Since June 1st 2023, I have been writing daily here on LinkedIn. My content reached 81+ Million impressions in the 12 months. Here are some reflections: 𝟭/ 𝗪𝗵𝗮𝘁 𝗜 𝗪𝗿𝗶𝘁𝗲 𝗔𝗯𝗼𝘂𝘁 I joined IBM over decade ago to work on AI and Machine Learning. Since then, I've had an incredible journey that I like to share here on LinkedIn, focusing on: - AI: I enjoy sharing real-world use cases, experiences, and my point of view on the latest market trends, as well as promoting our incredible IBM innovations. - Career and Leadership: From intern to executive in 8 years, I like to detail the keys that helped me navigate the corporate world, relocate multiple times, and lead large AI technical teams. - Personal Story: Having relocated from Europe to the US and lived in several countries, I hope my story inspires others. This includes how I manage parenting with ambitious professional aspirations. 𝟮/ 𝗛𝗼𝘄 𝗜 𝗪𝗿𝗶𝘁𝗲 𝘀𝗼 𝗠𝘂𝗰𝗵 I've always enjoyed sharing my learnings on blogs but never did it consistently. Many ask me how I manage to do it while working full-time with two small kids. Everything changed when I took Justin Welsh 's Content OS education. This helped me define my audience and create productive systems to scale. Here are some tips: - Idea Collection: If you're next to me, you'll see I'm always with a notebook in my pocket, constantly collecting ideas for topics. - Templates: I consistently analyze the best creators and create custom templates. - Reuse content: write once, publish ten times In a nutshell, I learned how to write and how to do it at scale. I also leverage AI Agents. If I were to get hit by a bus today, you would continue to hear from me until summer, given the automated systems I have in place and the content schedule I've set up. 𝟯/ 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆 I receive hundreds of DMs weekly here and many colleagues at IBM slacked me about my posts. From very senior executives to early-career professionals, I believe my writing here has helped: - Inspire people and educate them about AI - Recruit better talent for IBM - Unlock new business opportunities and enhance my credibility when talking to customers - Share with the world the incredible innovations my peers at IBM are bringing to market in the AI space. What brings me here every day are the enriching and thoughtful exchanges I have with all of you. THANK YOU to each of you who have engaged my content. Looking forward to our continued interactions! 🚀

Check out this fantastic Agents repo You will find an extensive and dynamic collection of agent tutorials and implementations. It is a great resource for: 1/ learning 2/ building 3/ sharing AI Agents ranging from simple conversational bots to complex, multi-agent systems. Key features: 🎓 Learn to build GenAI agents from beginner to advanced levels 🧠 Explore a wide range of agent architectures and applications 📚 Step-by-step tutorials and comprehensive documentation 🛠️ Practical, ready-to-use agent implementations 🌟 Regular updates with the latest advancements in GenAI 🤝 Share your own agent creations with the community Enjoy! Link here: https://lnkd.in/gnsbeztC ____ If you like this content, please repost it ♻️ and follow me, Armand Ruiz , for more similar posts.

Check out this fantastic RAG repo It comes with a comprehensive collection of advanced RAG techniques. A very valuable resource for researchers and practitioners seeking to enhance the 1. accuracy 2. efficiency 3. contextual richness of their RAG systems Key features: 🧠 State-of-the-art RAG enhancements 📚 Comprehensive documentation for each technique 🛠️ Practical implementation guidelines 🌟 Regular updates with the latest advancements Enjoy! Link here: https://lnkd.in/gz3ZtQuN ____ If you like this content, please repost it ♻️ and follow me, Armand Ruiz , for more similar posts.

This is a new game. It is not the Data Center game. It is the AI Factory Game. The more you spend. The more you save. The more you make. Good conversation in the Financial Analyst Q&A with Jensen at GTC.

🚀 Introducing BeeAI: an IBM Open Source platform designed to make working with AI Agents simpler, more flexible, and more powerful. Yesterday in San Francisco, some of my favorite colleagues from the IBM Research AI Incubation team presented the project's evolution at the AI Developer Conference organized by deeplearning ai. With BeeAI, whether you're building your own agents or leveraging existing ones, BeeAI helps you discover, run, and compose AI agents across any framework or language. A few highlights the team is especially proud of: 𝟭/ 𝗜𝘁 𝗶𝘀 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸-𝗮𝗴𝗻𝗼𝘀𝘁𝗶𝗰: Connect AI agents seamlessly, regardless of their language or platform. 𝟮/ 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝗱𝗲𝘀𝗶𝗴𝗻: Construct sophisticated multi-agent workflows using modular components. 𝟯/ 𝗕𝘂𝗶𝗹𝘁-𝗶𝗻 𝗱𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆: Explore a growing catalog of AI agents with an intuitive search. 𝟰/ 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿 𝘀𝘂𝗽𝗽𝗼𝗿𝘁: First-class tools for Python and TypeScript agent developers via the BeeAI Framework. And it's open source under the Apache 2.0 license. If you're working with agents or considering it, check out this out: - Website: https://beeai.dev/ - Documentation: https://lnkd.in/gqhy9d9c - Agent Library: https://beeai.dev/agents - Repo: https://lnkd.in/gyRGT8sA

The AI agent 'Manus' release is VERY impressive. Here are key use cases with demos. But first, 𝘄𝗵𝗮𝘁 𝗶𝘀 𝗠𝗮𝗻𝘂𝘀? It is an invite-only AI agent launched March 6 from China, excelling in research and execution, challenging Silicon Valley's AI dominance. 𝟭/ 𝗧𝗿𝗶𝗽 𝘁𝗼 𝗝𝗮𝗽𝗮𝗻 𝗶𝗻 𝗔𝗽𝗿𝗶𝗹: Creates personalized itineraries and a custom handbook. 👉 Demo: https://lnkd.in/g5SA3pjg 𝟮/ 𝗔𝗻𝗮𝗹𝘆𝘇𝗲 𝗧𝗲𝘀𝗹𝗮 𝗦𝘁𝗼𝗰𝗸𝘀: Provides in-depth stock analysis with dashboards. 👉 Demo: https://lnkd.in/g6vX2F8y 𝟯/ 𝗠𝗼𝗺𝗲𝗻𝘁𝘂𝗺 𝗧𝗵𝗲𝗼𝗿𝗲𝗺 𝗖𝗼𝘂𝗿𝘀𝗲: Develops video presentations for educators. 👉 Demo: https://lnkd.in/gGVQTXAV 𝟰/ 𝗜𝗻𝘀𝘂𝗿𝗮𝗻𝗰𝗲 𝗣𝗼𝗹𝗶𝗰𝘆 𝗖𝗼𝗺𝗽𝗮𝗿𝗶𝘀𝗼𝗻: Offers comparison tables with recommendations. 👉 Demo: https://lnkd.in/gPiiCBG3 𝟱/ 𝗕𝟮𝗕 𝗦𝘂𝗽𝗽𝗹𝗶𝗲𝗿 𝗦𝗼𝘂𝗿𝗰𝗶𝗻𝗴: Identifies suitable suppliers via research. 👉 Demo: https://lnkd.in/gVm6HBCq 𝟲/ 𝗔𝗜 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝘀 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵: Analyzes AI products in the clothing industry. 👉 Demo: https://lnkd.in/gFFbgdeN 𝟳/ 𝗬𝗖 𝗖𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀 𝗟𝗶𝘀𝘁: Compiles B2B companies from YC W25 into a table. 👉 Demo: https://lnkd.in/gpXG68Z4 𝟴/ 𝗢𝗻𝗹𝗶𝗻𝗲 𝗦𝘁𝗼𝗿𝗲 𝗔𝗻𝗮𝗹𝘆𝘀𝗶𝘀: Delivers insights from Amazon sales data. 👉 Demo: https://lnkd.in/g_UZyS6E 𝟵/ 𝗖𝗮𝗺𝗽𝗮𝗶𝗴𝗻 𝗠𝗮𝗽𝘀: Designs maps for historical events. 👉 Demo: https://lnkd.in/gWqTnMRY 𝟭𝟬/ 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝗦𝗰𝗵𝗲𝗱𝘂𝗹𝗲: Organizes interviews for 40 candidates. 👉 Demo: https://lnkd.in/gnVzWpUS 𝟭𝟭/ 𝗙𝗶𝗻𝗱 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿𝘀: Maps potential B2B customers. 👉 Demo: https://lnkd.in/gK2FKA-X 𝟭𝟮/ 𝗧𝗲𝗹𝗲𝗽𝗿𝗼𝗺𝗽𝘁𝗲𝗿: Creates a customizable teleprompter. 👉 Demo: https://lnkd.in/gzXbuHPA Link: https://manus.im/ ____ Please repost it ♻️ and follow me, Armand Ruiz, for more similar posts.

This project got over 49,000 stars on GitHub — check Langflow, an open-source, low-code app builder for RAG and multi-agent AI applications. Here's what you can achieve with Langflow: 🔧 Python-Based and Model-Agnostic: Build workflows that integrate seamlessly with any model, API, data source, or database. 🎛️ Visual IDE: Utilize a drag-and-drop interface to construct and test complex AI workflows effortlessly. ⚡ Playground for Immediate Testing: Instantly test and iterate on workflows with step-by-step control. 🤖 Multi-Agent Orchestration: Manage and coordinate multiple AI agents within your applications. ☁️ Free Cloud Service: Get started in minutes with no setup required. 🚀 Publish and Export: Deploy your applications as APIs or export them as Python applications. 🔍 Observability Integration: Monitor and optimize your workflows with tools like LangSmith, LangFuse, or LangWatch. 🔒 Enterprise-Grade Security: Benefit from robust security and scalability with the free DataStax Langflow cloud service. 🛠️ Customizable Workflows: Tailor workflows or create entirely new flows using Python. 🔗 Ecosystem Integrations: Access reusable components compatible with any model, API, or database. Langflow is available under the MIT license, and commercial versions are also available. Give it a try now: - Get started for free: https://www.langflow.org/ - Documentation: https://docs.langflow.org/ - Repository: https://lnkd.in/gSMce8-b We 💙 Open Source AI Happy building, folks!

BIG Announcement! IBM watsonx + NVIDIA NIM Integration, better together 💙💚 AI should be powerful, flexible and enterprise-ready. That’s why IBM is bringing NVIDIA Inference Microservices (NIMs) to watsonx.ai, making it easier than ever to build, scale, and deploy cutting-edge AI across cloud and on-prem environments. AI adoption is accelerating but complexity is holding many organizations back. With NIM integration, we’re simplifying enterprise AI deployment without compromising control, performance, or choice. 𝗪𝗵𝗮𝘁’𝘀 𝗮 𝗡𝗜𝗠? A NIM is a GPU-optimized AI model packaged in a container with built-in enterprise features — auth, monitoring, REST/gRPC APIs, and blazing-fast inference speeds. 𝗪𝗵𝗮𝘁 𝗲𝗻𝘁𝗲𝗿𝗽𝗿𝗶𝘀𝗲𝘀 𝗴𝗲𝘁: • Optimized performance on NVIDIA GPUs • Run AI anywhere with hybrid + multi-cloud support • Stronger AI governance with built-in security and observability • Faster deployment without vendor lock-in 𝗣𝗼𝘄𝗲𝗿𝗶𝗻𝗴 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 𝗮𝘁 𝗦𝗰𝗮𝗹𝗲 From smart assistants to complex workflows, agentic architectures rely on LLMs. NIMs give developers scalable, high-performance models with standardized APIs—accelerating how businesses build intelligent, real-time systems. 𝗦𝗲𝗮𝗺𝗹𝗲𝘀𝘀 𝗶𝗻 𝘄𝗮𝘁𝘀𝗼𝗻𝘅.𝗮𝗶 Imported NIMs appear right in your model library. Just select, prompt, and build—no extra steps. IBM’s vision is clear: open ecosystems, flexible deployment, and best-in-class models—from IBM, open source, or partners like NVIDIA. With watsonx, the future of enterprise AI is here. Learn more: - Official Press Release: https://lnkd.in/gtHydb9Q - Blog: https://lnkd.in/gr6HpMej

Forget RAG, welcome Agentic RAG 𝗡𝗮𝘁𝗶𝘃𝗲 𝗥𝗔𝗚 In Native RAG, the most common implementation nowadays, the user query is processed through a pipeline that includes retrieval, reranking, synthesis, and generation of a response. This process leverages retrieval and generation-based methods to provide accurate and contextually relevant answers. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗔𝗚 Agentic RAG is an advanced, agent-based approach to question answering over multiple documents in a coordinated manner. It involves comparing different documents, summarizing specific documents, or comparing various summaries. Agentic RAG is a flexible framework that supports complex tasks requiring planning, multi-step reasoning, tool use, and learning over time. 𝗞𝗲𝘆 𝗖𝗼𝗺𝗽𝗼𝗻𝗲𝗻𝘁𝘀 𝗮𝗻𝗱 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 - Document Agents: Each document is assigned a dedicated agent capable of answering questions and summarizing within its own document. - Meta-Agent: A top-level agent manages all the document agents, orchestrating their interactions and integrating their outputs to generate a coherent and comprehensive response. 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀 𝗮𝗻𝗱 𝗕𝗲𝗻𝗲𝗳𝗶𝘁𝘀 - Autonomy: Agents act independently to retrieve, process, and generate information. - Adaptability: The system can adjust strategies based on new data and changing contexts. - Proactivity: Agents can anticipate needs and take preemptive actions to achieve goals. Applications Agentic RAG is particularly useful in scenarios requiring thorough and nuanced information processing and decision-making. A few days ago, I discussed how the future of AI lies in AI Agents. RAG is currently the most popular use case, and with an agentic architecture, you will supercharge RAG!

Heard of MCP? Well, meet ACP! 😅 Our talented IBM Research team is tackling AI agent-to-agent communication challenges, and we've just published our initial draft of ACP (Agent Communication Protocol). 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗔𝗖𝗣? ACP standardizes agent interactions, enabling automation, collaboration, UI integration, and developer tools. Instead of enforcing strict rules upfront, ACP prioritizes practical features, ensuring adoption before formal standardization. 𝗠𝗼𝘁𝗶𝘃𝗮𝘁𝗶𝗼𝗻 Agent systems often face integration issues due to varied communication standards. ACP addresses this by streamlining interactions for agents that process natural language and rely on external models, fostering easier integration and collaboration. 𝗪𝗵𝘆 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗠𝗮𝘁𝘁𝗲𝗿𝘀 Standardization ensures reliable and widely adopted features. ACP evolves based on real-world use, offering: 1/ Interoperability: Smooth interaction between agents and systems. 2/ Simplified Development: Less complexity, more innovation. 3/ Efficiency: Reduced redundancy through reusable solutions. ACP also enhances agent-specific capabilities, such as: - Natural Language Flexibility: Optimized for handling ambiguous language. - Model Dependency Management: Simplifies external model integration and resource management. I invite you to join the discussion here or on GitHub about the ideas we're implementing: https://lnkd.in/gYkbbK5Y. Link to the initial draft: https://lnkd.in/gMeWegfb

Enterprise AI

Hottest Open Source project for Agents right now: MCP from Anthropic, and everyone should pay attention to it. The 𝗠𝗼𝗱𝗲𝗹 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹 (MCP) is set to become the new standard for connecting AI Agents & Assistants to the systems where data resides, including content repositories, business tools, and development environments. Its goal is to enable AI systems to produce more relevant and context-aware responses. 𝗧𝗵𝗲 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲: While AI models continue to advance in reasoning and quality, their capabilities are often constrained by limited access to data. Each new data source requires a custom integration, making truly connected systems difficult to scale. 𝗧𝗵𝗲 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: MCP provides a universal, open standard to securely connect AI tools with data sources. Instead of fragmented integrations, a single protocol simplifies development and ensures scalability, allowing AI systems to access the data they need efficiently. 𝗪𝗵𝗮𝘁 𝗠𝗖𝗣 𝗢𝗳𝗳𝗲𝗿𝘀: - A Specification and SDKs: Enabling developers to quickly implement MCP in their systems. - Local MCP Server Support: Integrated with Claude Desktop apps for seamless use. - An Open-Source Repository: Pre-built MCP servers for popular systems like Google Drive, Slack, GitHub, and Postgres. By standardizing the connection between AI systems and data sources, MCP enables AI tools to maintain context as they interact across different systems. This approach replaces fragmented integrations with a sustainable, scalable architecture. Learn more about the project and get started today here: https://lnkd.in/gi4BREN4 Github link to the MCP Servers repo: https://lnkd.in/gmYCvhST