Sandy Hardikar Interview
In this in-depth interview with Sandy Hardikar, he challenges a dominant assumption in enterprise transformation: that better tools automatically lead to better outcomes. Drawing more than two decades of experience, he explains that most organisations struggle not because they lack ERP, CRM, or AI systems, but because they lack a shared, living context that connects decisions, intent, policies, and institutional memory. The conversation highlights why AI first design delivers sustained ROI, how leadership must rethink AI governance, and why employee-led decisions are critical in an AI-first world. Ultimately, the interview frames shared context as the missing ingredient that allows organisations not just to adopt AI, but to evolve with it.
Real Reason for Enterprises Fail
We started the interview by asking, “Sandy, you often say enterprises don’t fail because they lack tools, but because they lack shared context. What do you mean by that?”
Sandy Hardikar replied, “Over the last two decades, I’ve seen organisations invest heavily in systems ERP, CRM, analytics, collaboration but still struggle to execute. The real problem isn’t technology scarcity; it’s context fragmentation. People can’t easily access the right information, don’t know why past decisions were made, or how policies, risks, and intent connect. Without shared context, AI can answer questions, but it can’t drive outcomes. Network Science exists to solve this exact gap by giving enterprises a living, compounding context layer that aligns decisions, actions, and accountability.”
Why AI Needs an Enterprise System
The UK Times: What is the “Enterprise System of Context,” and why is it so critical in an AI-first world?
Sandy Hardikar replied, “The Enterprise System of Context is the missing layer in enterprise architecture. Systems of record capture transactions, but they don’t preserve reasoning, intent, or decision history. In an AI-first world, that’s fatal because AI without context produces shallow value. Our system connects data, decisions, policies, workflows, and institutional memory into a single operating layer. This allows AI to move beyond answering questions to actually supporting, and in some cases driving enterprise decisions with governance, traceability, and confidence.”
Redefining Enterprise AI Beyond Automation
The UK Times: NSOffice.AI is positioned as more than a platform. How is it different from typical enterprise AI solutions?
Sandy Hardikar replied, “Most enterprise AI platforms optimise systems or automate tasks. NSOffice.AI is designed around how employees actually decide. We provide a single, secure interface (what we call One Decision Bar) to access every decision that is shaped by role, precedent, policy, and intent. Importantly, we sit above existing systems of record rather than replacing them.
So, when employees access particular information/content within the enterprise, they also get the complete context of why the document was created, why a section was edited and by whom, and so on. That’s how adoption of AI scales: employees are empowered, governance is preserved, and AI becomes a trusted business operating layer, not an experiment.”
How Context-First Design Delivers 8× ROI
The UK Times: Network Science reports an average 8× ROI across implementations. What drives that level of impact?
Sandy Hardikar replied, “ROI comes from clarity before automation. Through our AI Office engagements, we reimagine workflows within enterprises before deploying AI. We design context first, defining who decides what, with which constraints, and why. Only then we automate or augment. This prevents organisations from scaling broken processes. Over five years, across 150+ AI programs and 1,000+ workflows, this approach has consistently delivered measurable outcomes rather than pilot fatigue.”
Why Leaders Must Rethink AI as an Operating Layer
The UK Times: You work closely with CEOs, Boards, and policy leaders. How should leadership think differently about AI today?
Sandy Hardikar replied, “AI should not be treated as a technology initiative, it’s a business operating layer. Leaders need to ask: How does AI change how strategy is formed, how work gets done, and how value is measured? That requires governance, accountability, and enterprise memory by design. When AI is embedded into decision-making rather than bolted onto applications & workflows, organisations move faster with less risk. That’s the shift we help leadership teams make.”
Putting Employees Back at the Center of Decisions
The UK Times: Network Science emphasises “employee-led decisions, system-second design.” Why is that philosophy so important?
Sandy Hardikar replied, “Enterprises have been system-led for decades, forcing people to work around tools. We flip that model. Employees lead decisions; systems respond on demand. Empowerment isn’t about giving people more dashboards; it’s about giving them clarity at the moment of decision. When employees understand constraints, risks, and precedents in context, accountability naturally improves. AI becomes an enabler of judgment, not a replacement for it.”
The Future of AI Lies in Compounding Context
Lastly we asked, “Looking ahead, what is the single question driving your work at Network Science?”
“The question is simple but profound: How do we give enterprises a shared context layer that compounds over time? If we get that right, organisations don’t just adopt AI, they evolve with it. Decisions improve, execution accelerates, and value creation becomes continuous. That is the future we’re building at Network Science.” Sandy Hardikar concluded
Connect with Sandy Hardikar on LinkedIn
For more information visit Network Science Ltd.
Also Read:
Daouda Diabatè: Carrying Africa’s Living Voice through
Sorcha Drakeford, Founder of Design@Studio17
Tom Larsen: Building Value through Discovery
