Rosmon Sidhik

Building and Scaling AI Products in today's world is no easy job. You must serve hot masala dosas today while building the industrial kitchen that will feed millions tomorrow. Your product teams (pods) are the snack cart. agile, fast, and focused on immediate value. However the organization needs industrial Udupi kitchen: Built for quality, consistency, and scale. Both are essential. Leaning too far either way breaks the model. The Risks of only running Snack Carts : If we only operate snack carts, we feed customers immediately but breed chaos. - Multiple lentil suppliers : Fragmented model hosting and duplicated vector databases. - Inconsistent taste and quality : model drift, lack of evaluation discipline, and security gaps. - Secret batter recipes : Fine-tuning done in silos with no shared checkpoints or governance. - Sky-high spice costs : Uncontrolled API usage and unmanaged cloud spend. Speed without structure burns out teams and budgets alike. The Risks of only building Industrial Kitchens: If we only focus on the industrial kitchen, we’ll have the most sophisticated MLOps facility in town, serving no one. A perfect central kitchen gives you control but kills creativity. You trade speed for bureaucracy and risk building a scalable, empty restaurant while others feed the market. We sacrifice the very things that make AI products win "Perfection kills momentum. Over-engineering stalls learning". A Phased, Balanced Menu: Launch the Snack Cart (Next 3 to 6 Months) - Goal: Learn what dishes customers truly crave. Prioritise speed and customer validation. - Action: Use ready made batters: Cloud APIs, pre trained models, RAG. - Watch out: Do not mistake early excitement for product market fit. Build the Industrial Kitchen (Months 4 to 9) - Goal: Scale efficiently without losing flavour. Introduce versioning, observability, and governance while keeping innovation alive. - Action: Standardise core MLOps and data pipelines. - Watch out: Avoid process overreach that chokes experimentation. Scale the Udupi Chain (Month 10 and Beyond) - Goal: Deliver a complete, scalable AI experience. - Action: Integrate proven wins into end to end workflows with fine tuned models. - Watch out: Preserve agility even as systems mature. Guard fiercely against the bureaucracy and power struggles that often accompany scale. The Leadership Tension: We’re serving crispy dosas at the cart today while designing the kitchen for tomorrow. This friction isn’t dysfunction. It’s the rhythm of every modern AI company choosing to evolve rather than stagnate and every leader today faces this trade-off. Are you feeding customers today, or just stocking the pantry for tomorrow? #AI #TechLeadership #MachineLearning #MLOps #ProductManagement #ScalingAI #AIProducts #AIGovernance #EngineeringLeadership

20

Glad to have had the opportunity to speak on “The Deterministic AI Era – Re-Architecting Enterprise Infrastructure for LLM Workloads.” The conversations at the summit reinforced a strong belief — enterprise AI success will increasingly be defined by deterministic, production-grade infrastructure, not experimentation alone. As AI moves from pilots to production, engineering discipline, cost control, and architectural clarity become the true differentiators. Thank you Accion Labs for organizing a thoughtful and well-executed summit, and for curating a platform where current AI advancements were explored in depth. Truly appreciate the engaging discussions and diverse perspectives shared throughout the event. Looking forward to continuing the dialogue. #AccionLabs #InnovationSummit2026 #EnterpriseAI #LLMInfrastructure #AIEngineering

66

2 Comments

Came across this report from Melnoventures as to how consumers are using AI. Very interesting insights on the segmentation, use cases , opportunities. Out of the many insights few stood out "There are around 1.8 billion users globally who use AI tools . 1.8 billion users at an average monthly subscription cost of $20 per month equals $432 billion a year; today’s $12 billion market indicates that only about 3% pay for premium services—a strikingly low conversion rate and one of the largest and fastest-emerging monetisation gaps in recent consumer tech history. Even ChatGPT, with its first-mover advantage, only converts about 5% of its weekly active users into paying subscribers.This gap between usage and payment represents a major opportunity." We are in a phase where users are trying the tech out and getting used to it. Gradually the providers who are able to address the high usage, high friction , high trust use cases will be able to differentiate (have a moat) and charge a premium. In the current paradigm for consumers, switching costs are practically zero; there’s no data migration, no sunken cost. We’re looking for products that replace complexity with simplicity, embed persistent memory, have strong integrations, and unlock and train with proprietary user data. The AI providers are also trying to build a moat around these and once achieved they will control the consumers as has been the history with each technology pivot. Its an interesting marketplace where the major LLM players are also trying to build in specialised used cases as a part of their offerings. So the smaller players have to run fast and capture their relevant segment before they are run over by the big players. https://lnkd.in/g99ne5EC #GenAI #AIAdoption #AIopportunities #AIconsumers #AImonetization

7

Smarter Journeys Begin With AI in Retail Bots AI is everywhere in retail - but is it truly working for your brand, or just checking a box? At CrossML Pvt Ltd, we built a next‑gen AI Customer Support Bot that does more than talk: it thinks, learns, and sells - all while staying budget- and system-efficient. It’s turning support into sales, automation into loyalty, and casual chat into genuine growth. Why retailers - fast-growing and established - are betting on AI bots: -91% of retail IT leaders aim to make AI their top tech investment by 2026. -GenAI chatbots drove a 15% boost in Black Friday conversions . -60% of shoppers noticed better forecasting and inventory with AI retail tools . -By 2025, 85% of retail interactions will be fully automated. AI-powered support yields a 25% increase in customer satisfaction . 🛠️ How Our Bot Drives Smarter Journeys 1. Conversational AI that Feels Human & Converts Built with GPT‑4, Gemini & LLaMA - understands tone, intent, context, and nudges cart abandoners seamlessly. Impact: 63%+ retailers saw conversion gains, 55% saw higher average order value . 2. Seamless Backend Integration Connects Shopify, WooCommerce, CRMs, POS, loyalty & delivery APIs - provides real-time, personalized support. 3. RAG-Powered Real-Time Accuracy Fetches live data to answer so there’s never outdated or inaccurate info - drastically reduces hallucinations. 4. Empathy at Scale with Sentiment Analytics Our dashboard tracks neutral-to-negative sentiment and flags chats needing human attention - balancing speed with emotional intelligence. 🚀 Real-World Results ✅ 50% faster first responses during peak events ✅ 18% lift in daily sales through personalized recommendations ✅ 20% lower support operating costs ✅ 30% drop in repetitive queries ✅ 25% higher CSAT scores 🔮 What’s Next for Retail Bots - Voice & Visual Interactions - think kiosks, in-store voice support, and image-based assistance. True Omnichannel Presence - across live chat, email, WhatsApp, and more. Industry-Specific AI Modules - tailored to fashion, electronics, beauty, grocery - with domain logic built-in. Predictive Support - anticipating issues, identifying churn, and initiating outreach before customers even ask. At CrossML, our AI bots aren’t just hype - they’re built for real-world impact. If you’re a CTO, Founder, or AI Lead ready to elevate your retail strategy, let’s connect. Let’s make AI work smarter - for you and your customers. #RetailAI #CustomerExperience #ConversationalAI #GenAI #RAG #Omnichannel #CloudNative #CrossML

9

2 Comments

AI is shaping critical decisions in hiring, lending, and other sectors— and unintentional bias in these systems can have real-world consequences. I’ll be discussing how bias enters AI, the challenges of operationalizing fair models, and strategies to build more equitable solutions in an upcoming session with AnitaB.org. I invite fellow technologists, AI practitioners, and anyone passionate about responsible AI to join this conversation. 📅 Tuesday, March 17 | ⏰ 12 PM PT | 🔗 Register here: https://lnkd.in/emwMkAq2 #ResponsibleAI #AIethics #WomenInTech #DataLeadership #anitaborg

44