Jagan Nathan

Not getting time to catchup with the latest SF Data Cloud updates? Here’s a 60-second summary on the October 2025 Salesforce Data Cloud TRE: 🚀 𝗚𝗲𝗻𝗲𝗿𝗮𝗹 𝗣𝗹𝗮𝘁𝗳𝗼𝗿𝗺 • 𝗦𝗲𝗴𝗺𝗲𝗻𝘁 𝗰𝗿𝗲𝗮𝘁𝗶𝗼𝗻 𝗶𝗻 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗼𝗻 𝗼𝗿𝗴𝘀 – business units can now create and manage segments locally. • 𝗗𝗮𝘁𝗮 𝗖𝗹𝗼𝘂𝗱 𝗶𝗻 𝘀𝗮𝗻𝗱𝗯𝗼𝘅 – no more full refreshes; test safely with production licensing. 🔗 𝗖𝗼𝗻𝗻𝗲𝗰𝘁 & 𝗜𝗻𝗴𝗲𝘀𝘁 • 𝗠𝗮𝗿𝗸𝗲𝘁𝗶𝗻𝗴 𝗖𝗹𝗼𝘂𝗱 𝗖𝗼𝗻𝗻𝗲𝗰𝘁𝗼𝗿 𝘂𝗽𝗴𝗿𝗮𝗱𝗲𝘀 – now brings Journey Builder + Order data every 30 min with no extra credits. • 𝗡𝗲𝘄 𝘂𝗻𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲𝗱 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗼𝗿𝘀 – Guru, YouTube, GitHub (for RAG & enterprise search). • 𝗠𝗮𝗿𝗸𝗲𝘁𝗶𝗻𝗴 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗼𝗿𝘀 – Kakao Ads, Criteo Retail Ads, LinkedIn Company. • 𝗚𝗔𝟰 + 𝗚𝗧𝗠 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗼𝗿𝘀 𝗚𝗔 – setup time cut from weeks to minutes. ⚙️ 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺 & 𝗣𝗿𝗲𝗽 • 𝗟𝗟𝗠-𝗶𝗻𝗳𝘂𝘀𝗲𝗱 𝗥𝗔𝗚 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲 + 𝗗𝗼𝗰𝗸𝗹𝗶𝗻𝗴 𝗣𝗮𝗿𝘀𝗲𝗿 – smarter unstructured data parsing for PDF, DocX, PPTX. • 𝗗𝗮𝘁𝗮 𝗚𝗿𝗮𝗽𝗵 𝗳𝗶𝗹𝘁𝗲𝗿𝘀 – refine datasets, cut storage costs. • 𝗢𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲 𝗗𝗮𝘁𝗮 𝗖𝗹𝗼𝘂𝗱 𝗤𝘂𝗲𝗿𝘆 (𝗠𝗖𝗣) – natural-language queries → SQL. 🎯 𝗦𝗲𝗴𝗺𝗲𝗻𝘁 & 𝗔𝗰𝘁𝗶𝘃𝗮𝘁𝗲 • 𝗔𝗰𝘁𝗶𝘃𝗮𝘁𝗶𝗼𝗻-𝘁𝗿𝗶𝗴𝗴𝗲𝗿𝗲𝗱 𝗳𝗹𝗼𝘄𝘀 𝗚𝗔 – direct activations to any API endpoint. • 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗦𝗲𝗴𝗺𝗲𝗻𝘁𝘀 + 𝗕𝗿𝗼𝗮𝗱𝗰𝗮𝘀𝘁 𝗙𝗹𝗼𝘄𝘀 – real-time, parameterized audience sends. • 𝗚𝗿𝗼𝘂𝗽, 𝗥𝗮𝗻𝗸, 𝗟𝗶𝗺𝗶𝘁 𝗦𝗲𝗴𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 – fine-tuned audience control. • 𝗕𝗮𝘁𝗰𝗵 𝗗𝗠𝗢 𝗔𝗰𝘁𝗶𝘃𝗮𝘁𝗶𝗼𝗻𝘀 𝗚𝗔 – export entire DMOs to external targets. • 𝗖𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝘁 𝗳𝗶𝗲𝗹𝗱 𝗼𝗿𝗱𝗲𝗿 + 𝗥𝗲𝗺𝗼𝘁𝗲 𝗮𝗰𝘁𝗶𝘃𝗮𝘁𝗶𝗼𝗻 𝘀𝘆𝗻𝗰𝘀 for companion orgs. 💡 𝗔𝗰𝘁 • 𝗟𝗶𝗻𝗸𝗲𝗱𝗜𝗻 𝗜𝗻𝘀𝗶𝗴𝗵𝘁𝘀 + 𝗖𝗔𝗣𝗜 – near-real-time audience analytics + conversion tracking. • 𝗦𝗙𝗥 𝗠𝗶𝘀𝘁𝗿𝗮𝗹 𝗦𝗺𝗮𝗹𝗹 (𝟴𝗞) – improved embeddings and multi-turn retrieval. #Salesforce #DataCloud #MIDCAI #Sandbox #DataGraphs MIDCAI

27

5 Comments

Really great to see companies like Valo creating space for the community to come together, learn, and share what actually matters. Bringing admins, architects, and developers together to reflect on new releases, exchange perspectives, and learn from one another is how the ecosystem stays strong and grounded in real-world needs. Appreciate the leadership from Jari Salomaa and the Valo team for investing in community, conversation, and connection. This looks like a fantastic opportunity.

13

1 Comment

🔍 Function Calls vs Model Context Protocol (MCP) in Agentic AI 🤖 Function calling brought structure to LLM interactions — enabling models to trigger tools, retrieve data, and execute tasks. But in agentic systems, where LLMs act as thinking agents rather than reactive bots, function calls alone don’t cut it. Enter: Model Context Protocol (MCP) 🧠 🔹 Function Calls = “Here’s a tool. Use it.” 🔹 MCP = “Here’s context, memory, history, toolset, persona, and environment — go accomplish a goal.” Why MCP matters: • ✅ Context-rich reasoning — gives the model structured access to user goals, memory, history, and tool capabilities. • 🔄 Multi-turn continuity — allows agents to persist state and adapt plans over long tasks. • 🛠️ Tool orchestration — lets the model choose and combine tools intelligently rather than just calling one-off APIs. • 📦 Modularity & Portability — MCP creates a standardized, modular interface for managing agent inputs and outputs — beyond raw JSON and ad hoc hacks. Think of function calls like giving an intern a list of API docs. Think of MCP like onboarding a smart assistant with access to calendars, tasks, Slack, past conversations, and team goals — and letting them figure out what to do next. Agentic AI needs structure beyond prompts. MCP is that structure — enabling autonomy, reasoning, and reusable frameworks at scale. We’re moving from “chat with AI” to “collaborate with AI agents.” #AgenticAI #MCP #LLMs #AIInfrastructure #ToolUse #AIContext #AIArchitecture #GenerativeAI #FunctionCalling

39

2 Comments

The simultaneous emergence of Claude Cowork and Clawdbot (now rebranded as Moltbot) is a clear signal of where consumer AI is headed this year: genuine end-user productivity through semi-autonomous agents. The days of purely prompt-based, one-shot results are giving way to systems that can handle real, ongoing work based on high-level requests. These tools are starting simple, but their trajectory is unmistakable: they will likely evolve quickly into capable digital assistants running 24/7. If recent AI trends hold, most of us will soon dedicate at least one (if not several) always-on machines or cloud instances to handling our routine personal and professional chores. This shift could free up significant time and mental energy, allowing us to focus on higher-leverage activities with far greater returns. Security, however, remains the single biggest hurdle. Granting AI agents access to local files, email, calendars, or messaging apps introduces real risks such as prompt injection, data leakage, unintended deletions, or worse. While tools like Cowork use sandboxes and careful permissions, and self-hosted options like Moltbot keep data local, vulnerabilities persist and demand vigilance. For now, the wisest path is cautious experimentation: start small, isolate sensitive data, monitor closely, and contribute feedback to help shape safer, more reliable agents. These early steps are opening exciting new chapters for everyday users. AI, it appears, won't just answer questions, but will truly work alongside us.

36

3 Comments

Kicked off weekend following great conversation between Rowan Cheung & Satya Nadella about how AI is reshaping the future of work — from transforming software development to redefining knowledge work across industries like healthcare, finance, and science. 1. Tech’s True Impact vs. Hype Satya critiques society’s tendency to celebrate tech companies more than the actual impact of technology. Emphasis on shifting focus from admiration of companies to real-world outcomes, especially in sectors like education and healthcare. Case studies are insufficient—real transformation requires hands-on engagement. a) “I think we as a society celebrate tech companies far too much versus the impact of technology.” b) “Case studies don't help. You have to do it yourself.” 2. Human-Centered AI and Autonomy AI should augment, not replace, human decision-making. The concept of “human in the loop” is emphasized to counter overstatements about AI autonomy. Transparency and control are essential in agent-based systems. “Ultimately the human is in the loop, I think we overstate the autonomy here.” 3. Agentic Web and UI for AI Microsoft envisions a future where users manage fleets of AI agents through unified interfaces. The “UI for AI” is evolving—different modalities will serve different user needs. Integration across tools (e.g., Teams, Outlook, GitHub) is key to enabling this vision. “We are definitely trying to build a scaffolding for the AI age.” 4. Knowledge Work Transformation AI is reshaping workflows, making knowledge workers more empowered and employable. The traditional flow of information is being inverted—workers now prompt systems directly for insights. Encouragement to embrace tools and reshape work artifacts. “Use the tools, change the work.” 5. Developer Experience and AI-Generated Code AI is accelerating software development, addressing global tech debt. Tools like GitHub Copilot and coding agents are transforming how developers work. “Give me one tech intervention that makes a damn difference in education.” “We need a lot more software development… to work down that deficit.” 6. Enterprise Adaptation and Upskilling Companies must restructure around agents and embrace cultural reinvention. Upskilling is driven by tool diffusion, not just formal training. Personalized education and low-code tools empower employees to innovate locally. “The best defense against [displacement] is skilling and reskilling.” “It happened not by going to a training class… it happened by diffusion of the general-purpose tool.” 7. Proactive Agents and Real-World Value Proactive agents should interpret intent and act with minimal friction. Healthcare is highlighted as a domain where AI agents can deliver massive productivity gains. The goal is to make technology “disappear” into workflows. “Technology is powerful enough to disappear.” “The true value of AI is global economic growth.” https://lnkd.in/edEcC5E5 #futureofwork #agenticAI #copilot

8

AI agents are reshaping how we build applications by enabling flexible, goal-oriented systems that can reason, act independently, and adapt to real-time data. Unlike traditional rule-based apps, AI agents leverage large language models to make nuanced decisions, integrate with diverse APIs, and respond to personalized user preferences—all while reducing development time and complexity. This shift unlocks new possibilities for personalized experiences, faster prototyping, and more resilient integrations, but it also introduces challenges around predictability, stability, and operational monitoring as these agents move into production environments. For data and analytics professionals, the rise of AI agents means more than just automation—it’s about creating systems that proactively deliver insights, optimize workflows, and interact naturally with users. As Google Cloud and others roll out robust agent-building frameworks and tools, now is the time to explore how agentic applications can take your data-driven solutions to the next level. Dive deeper into the practicalities and benefits in this excellent overview: https://lnkd.in/gYXnha8j

5

Scaling inference for LLMs presents a unique set of challenges, very different from how we've historically scaled traditional applications. This is an amazing post by the llm-d contributors showing how they're tackling this head-on. They detail a smart, AI-aware scheduling approach that considers both KV-cache hits and dynamic server load to dramatically boost performance and reliability.

15

1 Comment

🎙️ Co-presented at the Dynatrace India Monthly Virtual Tech Talk with Meghana Shroff — and this one hit close to home. The session: Mastering Observability: Best Practices for Peak Performance We opened with a story about Reena — a pregnant woman who couldn't submit a healthcare claim because the portal was broken. Nobody in the backend team knew. That's not a tech problem. That's an observability culture problem. Here's the gist of what we unpacked The mistakes we see way too often: → Alert storms from static thresholds — your team is numb to them by week 2 → Ingesting tons of data but zero business context → Dashboards that look cool but tell you nothing actionable → L2/L3 teams only showing up after production blows up → Missing tags = missed alerts. Simple. Brutal. → Config debt silently growing because nobody audits What actually works: → Let Davis AI learn seasonal baselines — stop manually setting thresholds → Alert on failure rate, not failure count → Enrich logs with trace context — no more timestamp-matching across 5 tools → Track user journeys end-to-end: login → search → payment → booking → Use forecast analyzer to catch resource exhaustion before it happens → Shift from firefighting → prevention and self-healing The closing line that stuck with me: "Observability is about people, not just platforms. Outcomes, not just metrics. Prevention, not just firefighting." Reena's issue? High CPU on one host during a predictable peak. Completely preventable. The team had all the data — they just weren't using it right. Thanks to everyone who attended, asked sharp questions, and made this genuinely fun to deliver. On-demand recording: https://lnkd.in/gSjjbhWG #Dynatrace #Observability #SRE #DevOps #PlatformEngineering #Davis #RUM #OpenTelemetry #BestPractices #India #TechTalk Special Mentions and thanks for all the support we got, Resha Raut, Chloe He, Dhanarajasekaran Kuppusaami, Darshan Desai, Prasad Khamkar, Nitin Yadav, Jaydeep Dubey, Natalie Barrett, Jitendra Sharma

21

🚀 Dive Deeper into the Future of AI Agents: Understanding the Model Context Protocol (MCP) The buzz around Agentic AI is undeniable. These intelligent systems are poised to revolutionize how we work, create, and interact with technology by taking autonomous actions and making decisions. But what's the foundational plumbing that will allow these agents to seamlessly connect with the world's vast digital landscape? A key part of the answer lies in the Model Context Protocol (MCP). If you're looking to get a clear understanding of what MCP is, its goals, and how it's set to standardize the way AI agents interact with tools and data, the official introduction is the perfect place to start. ➡️ Explore the fundamentals here: https://lnkd.in/gEB_5N3k Understanding MCP is crucial for anyone involved in AI development, product strategy, or simply curious about the next wave of AI innovation. This protocol aims to: 1. Simplify Integrations: Reduce the complexity of connecting AI agents to diverse systems. 2. Boost Agent Capabilities: Enable agents to access and utilize a wider array of tools and context. 3. Foster an Open Ecosystem: Encourage interoperability and collaboration in the development of agentic systems. By creating a common language for AI agents and the services they interact with, MCP is laying the groundwork for more powerful, versatile, and ultimately, more helpful AI. Take a few minutes to explore the introduction and get up to speed on this pivotal development. What are your initial thoughts or questions after reading it? Let's discuss how MCP will unlock new possibilities for AI! Share your perspectives in the comments. 👇 #ModelContextProtocol #MCP #AgenticAI #AI #ArtificialIntelligence #OpenStandards #FutureTech #Innovation #LLMs #AISystems #TechCommunity

131

32 Comments