Karan Talati

Ram Rajya is back. This year at NVIDIA GTC, after all the flashy launches, one thing became very clear to me: We’re no longer paying for compute. We’re paying a heavy premium for memory. GPUs are becoming commoditized faster than we expected. TFLOPs are everywhere. But RAM is the real bottleneck. LLMs don’t choke because of lack of compute. They choke because context needs memory. Agents need memory. Inference at scale needs memory. The future isn’t CPU-bound. It isn’t even GPU-bound. It’s RAM-bound. Whoever controls: high-bandwidth memory larger context windows cheaper, scalable RAM access …controls the next wave of AI. We’re quietly moving into a world where: Memory is the new oil. Welcome to Ram Rajya 😅 And this time, it’s not a mythological one.

205

7 Comments

✨ Wrapping up IBC & RDK Summit – Insights, Lessons, and What’s Next The past week at IBC and the RDK Summit have been nothing short of transformative — from big ideas to actionable strategies. 🔑 Key Takeaways At IBC, conversations revolved around AI-driven workflows, monetization models, and creating differentiated user journeys. At the RDK Summit, the emphasis was on making RDK operator-friendly — ensuring faster deployments, simplified integrations, and consistency across devices. 📚 Lessons Learned Storytelling and use cases matter more than just technology — whether it’s sports enhancements that redefine live experiences, or Kids OTT platforms built for safe, engaging entertainment. Operators are looking for practical innovation — solutions that are easy to adopt, deliver ROI, and scale with user expectations. The future of media operations lies in AI-powered management, ensuring agility, automation, and efficiency at every step. 🚀 Hero Solutions We’re Driving Forward - Sports Enhancements – smarter fan experiences powered by real-time insights. - Kids OTT – safe, fun, and personalized platforms for the next generation of viewers. - AI-based Managed Media Operations – intelligent automation that reduces complexity and accelerates innovation. ⏭️ Next Steps - Strengthen collaboration with operators and partners to drive operator-friendly RDK deployments. - Keep building narratives around real-world use cases rather than just technical features. - Translate the momentum from IBC - International Broadcasting Convention & RDK Summit into tangible solutions for the industry. A big thank you to everyone who engaged with us — your ideas and energy are helping shape the next chapter of media and connected TV. #IBC2025 #RDKSummit #SportsTech #KidsOTT #AI #ManagedMedia #InnovationsinMedia #LTTS Atul Kapur Jitesh Kumar Popat Anshuman Singh Ruben Thalakkottur Joseph Ajith Prasannan [Ajith] Andy Gangotra Rajarajan A Lakshmikanthan (RR) Paddy Viswanadhula Ruuchi Padhi Ashish Moghe Bhavna Oberooi

28

Project Prometheus from Jeff Bezos raising 100 Billion to bring AI automation to Manufaturing industry. https://lnkd.in/eDWAm82r While Project Prometheus reimagines the factory floor, I am collaborating on a fund to transform the Suncoast Belt's home services sector. We aren't just adding software to old models; we are building multiple AI-native companies designed to eliminate the friction that defines the industry today. At its heart, this is about the human experience. Homeowners shouldn't have to deal with the anxiety of "no-call, no-shows" or the frustration of misdiagnosed repairs that leave their families in the heat. By automating the logistical chaos and reducing human error through AI, we restore trust and peace of mind to the customer. For the technicians, we remove the burnout of inefficient routing and paperwork, allowing them to focus on the craft of service. We are scaling a more empathetic, reliable, and efficient way to care for the places people live, one route at a time

5

Scaling Development Across 250+ Engineers: Bitso started with a single shared dev environment and a single shared staging environment. Devs often tested a collection of changes from multiple engineers in staging, then deployed a different build to production, meaning the artifact that reached customers had never truly been tested in a clean environment. Check out how Joe Horsnell and his team fixed these issues by building RISE (Robust Isolated Staging Environment) using Signadot for service isolation Neon for state isolation. Now: - Every engineer gets a production-like environment by default - Changes are tested independently and safely - Deployments become predictable and repeatable - Staging stability no longer controls team velocity https://lnkd.in/ecXWC_G5

49

2 Comments

Robotics isn’t about a hardware. It’s a reliability, security, and financing story. In my conversation with Anand V Lalwani from Cardinal Robotics, one thing was clear: the winners in physical AI may not be the companies building the robots, but the ones making them deployable at scale. And that’s the space Cardinal is building in. They don’t just roll robots onto a site, they: → finance them (solving the RaaS cash-flow gap) → secure them (a requirement for pharma, semis, defense, and Fortune 500) → maintain fleet uptime nationwide → layer in a human workforce that operates alongside the machines That’s how pilots turn into multi-site enterprise contracts. Two takeaways that stuck with me: 1. In robotics, uptime is the product. 2. If your customer’s business model isn’t proven yet, you’re taking risk squared, a useful lens for parts of today’s humanoid and training-data rush. Their 2026 focus is just as interesting: using AI to train and deploy disabled veterans as on-demand field technicians, a new labor model for a robotics economy. The real value in physical AI will be built in the infrastructure layer. If you’re building in physical AI, are you a product, or the infrastructure every product will depend on?

71

3 Comments

Great time at the Liquidity Summit two weeks ago. I had the chance to sit down with the NYSE right after Bill Ackman to talk AI investment trends, compute demand, and what's driving the next phase of growth. Here's the conversation.

93

1 Comment

I love have these discussions ... errr debates ... err arguments with Radhika Malik, except for the fact I usually lose. To be sure, moving up the stack and moving down the stack are both hard problems. And interestingly, there are ebbs and flows in this regard. What I mean is, certain macro economic situations drive this behavior. In today's environment, the massive supply-demand imbalance is driving the inference providers to think about securing supply to support their growth from demand. Similarly, the neo-clouds and other infrastructure providers are seeing the huge inference demand and salivating at the additional revenue (and likely the improved margins) as well as the direct customer relationship they could capture by moving up the stack. I'm artfully (still) dodging the question ... so who has the easier path? As I said, both hard, but if you are a neo-cloud or infrastructure provider, you've already solved the hardest problem (to some extent anyway), which is securing allocation of GPUs (or other AI infra). Creating a performant inference stack is hard, but with vLLM, SGLang, TensorRT and other open source options available, constructing an inference stack seems like the easier path. Even if they end up building something not as good as the really great inference engines from the likes of a FAL AI, Fireworks AI, etc, the current demand equation will soak up everything they have to offer. And over time, that usage (knowledge) should only help them improve their inference product and close the gap with the leaders. Conversely, the inference providers need to build out infrastructure. That's real world sh#$! Securing data center / colo space, getting power (!!!), setting up gear (racking and stacking), configuring clusters and networks, securing allocation of GPUs (yes, they probably already have this partially managed), ... while not rocket science, takes hard work, unique skills, and frankly long-ish lead times. Right Zachary Smith? But perhaps there is a corollary question to ask. When (that's right "when" not "if") the supply-demand imbalance environment (finally) changes, which of these players will find themselves in the best place to adapt and thrive in that environment? Let's see what Radhika Malik has to say about that.

3

1 Comment

🚀 BIG NEWS in AI! Indian-origin company Uniphore just raised a massive $260 million (Series F) from global tech giants NVIDIA, AMD, Snowflake, and Databricks — pushing its total funding close to $870 million! 💰 Founded by Umesh Sachdev and Ravi Saraogi, Uniphore has become one of the world’s leading Enterprise AI platforms — seamlessly blending Conversational AI, Emotion AI, and Generative AI for businesses worldwide. 🌍 💡 Why this matters: 🔹 Enterprise AI is no longer experimental — it’s going mainstream. 🔹 Backed by chip, cloud, and data-analytics giants, Uniphore is poised to redefine how organizations automate customer and employee interactions. 🔹 A proud moment for Indian innovation making global impact! 🇮🇳 🔥 Key Investors: ✅ NVIDIA ✅ AMD ✅ Snowflake ✅ Databricks 💬 What this signals: The future of business AI will be agentic, composable, and deeply integrated into every workflow — and Uniphore is leading that transformation. 👏 Congratulations to the entire Uniphore team, Umesh Sachdev, and Ravi Saraogi for putting India on the global AI map once again! 👉 Read the full story here: https://lnkd.in/gbsWfg-E #Uniphore #AI #EnterpriseAI #StartupFunding #Innovation #NVIDIA #AMD #Snowflake #Databricks #TechNews #ScoopEarth #IndianStartups #UmeshSachdev #RaviSaraogi #scoopearthmagazine

12