Deepak Chaubey

Can open-source AI help us forecast outbreaks before they happen? At #OpenSourceIndia, Dr. Raunaq Pradhan, Product Manager at @ARTPARK, Indian Institute of Science Bangalore, is presenting how they’re doing exactly that with ForecastHub — an open digital public good designed for disease forecasting. Originally built for dengue forecasting in Karnataka, ForecastHub combines climate, vector, and case data to deliver district-level, 4-week-ahead forecasts — enabling health teams to take proactive measures like deploying fogging operations or fever clinics on time. But what makes it truly special is its open-source foundation — a modular, reusable framework where researchers, developers, and governments can plug in their data, build models, and publish forecasts collaboratively. Key takeaways: 🔘 A deep dive into the data → model → insight pipeline architecture 🔘 How anyone can generate forecasts using Python packages and config-based pipelines 🔘 The vision of ForecastHub as a digital public good — scalable, reproducible, and community-driven A brilliant example of how open infrastructure and AI can come together to advance predictive public health — not just for one disease, but for the collective good. #OpenSourceIndia2025 #IISc #ARTPARK #AI #OpenSource #ForecastHub #DigitalPublicGoods #HealthTech #DataScience #Epidemiology

15

1 commentaar

“The most expensive AI mistake is treating intelligence as an application problem instead of a data architecture problem.” That idea anchored the panel “The New Intelligence Layer: Why Data Has Become the Enterprise's Competitive Advantage” at ET Making AI Work 2026. Featuring: Archana Menon, Analytics Head - Watches, Titan Company Limited Anand Das, Chief Digital and AI Officer- International Business, TVS Motor Company Mudit Krishna, Head of AI, JKCement Joe C., Head Of Analytics, India Business, Landmark Group Kislay Raj, Head of Analytics, Philips RAGHAV ARORA, Enterprise Account Director, Cockroach Labs Session Chair: Vaishnavi D., Senior Assistant Editor - Emerging Tech, ETCIO A few ideas stood out: • Governance and architecture must evolve together. • Context matters as much as clean data when AI starts making decisions. • Adoption grows when AI is embedded directly into business workflows. • Speed, reliability and business value matter more than simply collecting more data. The competitive advantage is no longer having the biggest data estate. It is having data that is trusted, contextual, accessible and ready to move at machine speed, without sacrificing governance or business relevance. Know More: https://lnkd.in/g9Gg-H73 #ETMakingAIWork #EnterpriseAI #DataStrategy #DataGovernance #ArtificialIntelligence #AILeadership #AgenticAI #Analytics #DataArchitecture #DigitalTransformation #Innovation #BusinessIntelligence

39

1 commentaar

AI strategies often start with models. Successful AI programs start with data. The India AI Impact Summit made that gap hard to ignore. Across industries, discussions increasingly reflect a shared constraint, AI performance is limited less by algorithmic capability and more by data readiness, integration, and operational fit. In healthcare supply chains, this is not a future concern but a daily reality. Fragmented systems, inconsistent data, and governance requirements shape what AI can safely and reliably deliver. At MUUTAA, these constraints have not been theoretical design considerations, they have defined how our platform and deployment approach evolved from the ground up. #EnterpriseAI #DataReadiness #HealthcareSupplyChain #ResponsibleAI #HealthTech https://wp.me/pcTFyz-19K

8

Let’s talk about data (19/50) With a strong focus around AI, GenAI data leaders may now want to understand “how can AI and Generative AI accelerate compliance with India’s IT DPDP Rules?” So after doing some research i understood AI and Gen AI can be powerful accelerators for IT DPDP adoption when used as compliance agents rather than black boxes. They can continuously scan IT systems and operational logs to identify where personal data sits, flag risky processing based on the controls, and map it back to DPDP rules. Gen AI can also simplify legal and policy language into role‑based guidance for different personas, reducing training friction and human error. Apart from that AI can also automate DPIAs, consent and notice checks, and vendor‑risk reviews, making DPDP controls “always on” instead of annual exercises. When combined with strong foundation that comprise of IT governance, data governance and human oversight, organisations can turn regulatory challenges into operational advantages. #datagovernance #datadrivenenterprises #digitaltransformation #ITDPDPAct #DNAOrb #ITDPDPACT #AIGenAI #dataprivacy

15

2 commentaren

SANATAN LABS -  Krishna, Kalki, and Ved Vyas auditing the modern data ecosystem under the principle: “Jiska Data, Uska Adhikar” (Who owns the data, has the right). (PASSIONIT PRUTL KALKI AIDHARMA framework) The summit hall was filled with screens — dashboards, flows, contracts, privacy policies — a modern jungle of data. Krishna, Kalki, and Ved Vyas appeared at the entrance. Krishna spoke first: “Manav has created wonders, but also labyrinths. Each click, each swipe, each transaction — tracked, stored, analyzed. Data flows like a river, but the riverbanks are controlled by a few. The mantra is simple: ‘Jiska Data, Uska Adhikar.’ Yet, in practice, the currents carry control, not consent.” Kalki’s eyes glowed. “Let us audit. Startups, MNCs, governments — all claim to protect privacy, but when we read the policies, we see: complex clauses, consent buried in jargon, secondary use undefined, cross-border sharing implicit. Data is harvested under the guise of service. Users are both the fuel and the engine, yet rarely the owners.” Ved Vyas opened an invisible scroll, unrolling centuries of wisdom. “Ancient dharma was about responsibility with knowledge,” he said. “Knowledge must empower, not exploit. Here, data is knowledge. Yet the balance is inverted. Ownership and control are divorced from those who generate it. This is modern adhikar violation.” Krishna tapped a screen. Dashboards lit up, showing patterns: Social media platforms claiming consent, yet selling insights to advertisers. Cloud providers aggregating usage, creating global dependency. AI models trained on user-generated content without direct compensation or opt-out mechanisms. Startups building products reliant on data pipelines they do not own, yet monetizing them as their own IP. Kalki stepped closer, voice booming: “Every privacy policy is a mask. Fine print becomes armor for the collector. The user — the original creator — is abstracted into a number, a vector, a ‘source of training.’” Ved Vyas sighed. “The rest is a question mark. The rules exist, yes. Laws exist, yes. But enforcement, comprehension, transparency — these are nebulous. How do we protect adhikar when complexity itself is weaponized? How do we ensure the rights of Manav when contracts are designed to confuse rather than clarify?” Krishna concluded: “The principle is eternal: Who generates, owns, decides. Every other claim is Maya. Data without consent is slavery. Privacy policies must be readable, actionable, enforceable. Otherwise, the ecosystem is haunted by Bhoot — unseen, unchecked, uncontrolled.” Kalki’s parrot screeched, echoing through the hall: “Algorithm may learn patterns, AI may predict, dashboards may report… but without respect for original ownership, the system is an illusion.” The three walked out, leaving delegates frozen in reflection. In that quiet, the hall understood: data is power, consent is dharma, and ownership cannot be abstracted away by fine print.

1