For decades, technology distribution ran on transactions: an order, fulfilment, credit, collection, done. That model is running out of road. As Indian enterprises adopt AI, cloud and cybersecurity together, they now expect a partner who stays accountable long after delivery. Redington Limited is reinventing itself as a technology orchestrator, building platforms such as CloudQuarks and AI Exchange that offer over 450 ready-to-deploy AI agents, and investing in skills at scale. In conversation with CISO Forum, Deepak Puligadda, Global CTO at Redington Limited, explains why data readiness matters more than GPUs, how governance and skills must move together, which sectors will scale AI fastest and which will lag, and what the wider channel must change to avoid being left behind.
CISO Forum: Redington has traditionally been known as a technology distributor. Walk us through what “technology orchestration” means in practice. What capabilities had to be built that didn’t exist in the old distribution model?
Deepak Puligadda: At its core, our role has always been to bridge the gap between innovation and adoption. What has changed is the engagement model, value creation and the enablement approach.
The traditional distribution model was designed around transactions – an order, fulfilment, credit, collection, and the engagement largely closed there. Orchestration is designed around the lifecycle. It means building platforms, capabilities, and ecosystem connections through which technology is discovered, deployed and adopted, with the customer realising value at every stage.
To do that, we had to build capabilities that did not exist in the old model: digital platforms with API-first integration to vendors, marketplaces for cloud and AI solutions, customer success and lifecycle management functions, automated renewals, consumption analytics, and a services capability across cloud, security, and AI. We also had to invest in skill-building at scale for our own teams and for partners.
In that sense, distribution moved products. Orchestration enabled outcomes.
CISO Forum: India is often described as shifting from AI pilots to production-scale deployments. From where you sit across hundreds of enterprise customers, what’s actually driving that shift: cost, competitive pressure, or something else?
Deepak Puligadda: I don’t think it is a single factor. There are many things at play, but I will call out a few important ones.
First, use cases have matured. After 18 to 24 months of experimentation, organisations now know where AI genuinely pays back – service operations, document processing, forecasting, code assistance – and that clarity itself accelerates commitment. Second, time-to-value has compressed. Solutions that once required months of custom development can now be found, configured, and deployed in weeks through pre-validated agents and marketplaces. Third, competitive pressure has shifted from “should we do AI?” to “how fast can we scale it?”, and that question now comes from the boardroom.
Cost is more an enabler than a driver; cloud has democratised access to high-end compute. The honest caveat is that the shift is real but uneven. Organisations are not struggling to find AI solutions; they are struggling to identify the right use cases, integrate AI into existing workflows, and scale responsibly.
CISO Forum: Most enterprises today are adopting AI, cloud and cybersecurity simultaneously rather than sequentially. How has that changed what enterprises expect from a technology partner, and where do partners typically fall short?
Deepak Puligadda: Enterprises don’t experience these as separate entities. The moment an AI initiative is approved, it immediately raises questions of cloud capacity, data security, identity, and compliance. So, expectations have shifted from procurement to accountability: customers want one partner who bundles everything together, ensures seamless deployment, simplifies the engagement, and stays present after the transaction through services.
Where partners typically fall short is in operating in silos – a cloud practice here, a security practice there, with little integration between them. Many still sell products when customers ask for an outcome. And too often, engagement ends at delivery rather than continuing through adoption and optimisation. The partners who will succeed are those who bring connected, cross-domain thinking and who treat the transaction as the beginning of the relationship, not the end.
CISO Forum: What does the infrastructure layer compute, storage, networking need to look like for an enterprise serious about scaling AI beyond pilots, and where are most Indian enterprises underinvesting today?
Deepak Puligadda: I would start with data, not compute. GPUs attract attention, but data readiness determines whether AI scales – quality, pipelines, accessibility, and governance. That is where I see the most consistent underinvestment among Indian enterprises.
Beyond that, a few things matter. First, architectural flexibility: modular, vendor-agnostic stacks that can pivot as models, pricing, and regulation evolve. Second, a hybrid view of compute: cloud for elasticity, on-premises and edge where latency, sovereignty, or economics demand it. Third is financial discipline. AI consumption costs change quickly, and without visibility and FinOps practices, scaling becomes expensive experimentation. Fourth is security designed for AI workloads from the outset, rather than retrofitted later.
The right mix of these factors enables the development of the right foundations and operating models that determine how and when pilots succeed.
CISO Forum: Skills and governance are often cited as blockers to AI adoption. In your experience, which is the bigger bottleneck for Indian enterprises right now, and what would meaningfully move the needle on either?
Deepak Puligadda: The skills gap is often misunderstood. It is not only about data scientists. The real shortage is in people who can translate a business problem into an AI use case, and in partners who can implement, integrate, and support solutions reliably. What meaningfully moves the needle is structured enablement at scale through certifications, practitioner programmes, hands-on validation environments, and learning by doing rather than learning in theory. This is why we have invested in Redington Academy and our AI Centres of Excellence, integrating OEMs, ISVs and partners to benefit from these and contribute to capability building across. We believe no single organisation can build these capabilities alone.
On governance, my advice is simple: don’t treat it as an afterthought, but don’t wait for perfect frameworks either. Lightweight, responsible-AI guardrails adopted early scale far better than heavy controls imposed late. Skills without governance create risk; governance without skills creates paperwork. Enterprises need both, sequenced sensibly.
CISO Forum: Tell us about CloudQuarks and AI Exchange: what specific adoption friction were these platforms built to solve, and how do you measure whether they’re actually working for customers?
Deepak Puligadda: Both platforms were built to remove specific friction, not to add another layer of technology. CloudQuarks addresses the operational complexity of running a cloud business across multiple hyperscalers, such as fragmented procurement, subscription sprawl, billing complexity, and poor cost visibility. It brings discovery, procurement, provisioning, subscription management, renewals, consumption analytics, and cost optimisation into one unified experience, and increasingly serves as a gateway to AI-ready infrastructure.
AI Exchange addresses the distance between interest and implementation. Customers were interested in AI but couldn’t easily find validated solutions; ISVs and partners built innovative solutions but couldn’t find the right opportunity to deploy them. Today the platform offers more than 450 curated, ready-to-deploy AI agents across industries and business functions. These are pre-validated solutions that can be configured and deployed in weeks rather than built from scratch over months. We measure success through adoption, production conversion, and expansion rates.
CISO Forum: AI Centres of Excellence have become a common structure inside large enterprises. What separates a CoE that successfully scales use cases from one that stalls out after a few proofs of concept?
Deepak Puligadda: The CoEs that stall usually share a pattern: proofs of concept without a business owner, no defined path to production, and success measured in demonstrations rather than outcomes. They become showcases rather than engines.
The ones that scale are anchored to a business problem with an accountable owner and a defined metric. They run short validation cycles and scale fast. They design the path to production from day one, including integration, change management, funding, etc. They also help with capability building, so that expertise spreads across the business units rather than remaining locked inside the CoE.
A well-run CoE should, over time, make itself less necessary because the organisation itself becomes capable.
CISO Forum: As AI adoption matures, how is the profile of a “technology partner” changing? Are enterprises now expecting orchestrators like Redington to take on advisory or even outcome-based accountability, rather than just supply capability?
Deepak Puligadda: The whole ecosystem is evolving, and enterprises are increasingly looking for a partner to advise on use-case selection, architecture, and cost models; to remain engaged across the lifecycle from onboarding to renewal;
An orchestrator like us brings it all together to reduce time-to-value and ensure solutions are delivered and outcomes are achieved.
CISO Forum: Looking two to three years out, which sectors do you expect to move fastest on enterprise AI adoption in India, and which are likely to lag and why?
Deepak Puligadda: The pace of change makes predictions genuinely hard, but directionally a few patterns are clear. BFSI will lead. Fraud, risk, and customer service automation are mature use cases, and the sector has the data estates to support them. Retail and e-commerce will move quickly on personalisation and supply chain intelligence. IT services and Global Capability Centres are both adopters and force multipliers, and healthcare is gaining ground in diagnostics and operations. Manufacturing will see selective leaders, particularly where edge AI and quality inspection use cases are strong.
Certain sectors, such as MSMEs and cost-sensitive segments, will take longer to scale. Data readiness and affordability are critical here. That is precisely where ecosystem models matter most, providing access to pre-validated solutions delivered through trusted local partners.
CISO Forum: For India’s broader technology channel ecosystem distributors, resellers, systems integrators what has to change structurally for the industry to keep pace with enterprise demand, and who do you see being left behind if they don’t adapt?
Deepak Puligadda: We have already started witnessing a shift from one-time transactions to lifecycle-led models. The economics of this industry are driven by adoption, renewal, and expansion. Now what needs to happen is a steady transition from product resale to capability-led selling, with services attached to every solution.
From a Distributor’s standpoint, it means investing in platforms, skills, and ecosystems, becoming orchestrators rather than intermediaries. That’s where Redington has started focusing and investing now.
I am optimistic about this transition. The demand is real and growing, and partners don’t have to make this shift alone. Enabling that transition is, in many ways, the reason Redington exists in its new form.
