In an engaging conversation with Express Computer, Sagar PV, Chief Technology Officer, Mindsprint, outlines the company’s bold evolution from a captive tech arm of the Olam Group into a full-fledged, IP-led IT services firm. With over 95% of its innovation and delivery rooted in India, Mindsprint has rapidly scaled its GenAI and Agentic AI initiatives, aiming to lead with pre-built, AI-enhanced solutions across industries like food and agri, retail, manufacturing, and life sciences. Sagar details how Mindsprint’s “AI-first” and “future-forward” vision is not only reshaping its delivery models but also driving outcome-based engagements, experimentation in quantum computing, and strategic expansion into global markets.
Can you tell us about Mindsprint’s operations, particularly your India presence, and how you’re positioned as part of the Olam Group?
We are an offshoot of the Olam Group, originally functioning as its technology arm. In 2022, we became an independent entity, though still governed by the Olam board, with our own leadership, sales, P&L, and profitability structure under the Mindsprint name.
Our main delivery centres are in Chennai and Bengaluru, with global presence in Singapore, Western Africa, London, the Middle East, and the US. While global teams handle sales and delivery at client sites, 95% of our delivery, innovation, and R&D is based in India.
We’re a rare example of a GCC successfully transforming into a profit-focused IT services company. Over the past two years, we’ve added nearly 30 new clients beyond Olam, broadening our industry and geographic footprint. While our roots are in food and agriculture, we now serve clients across manufacturing, life sciences, retail, CPG, energy, and banking. However, our core focus remains on food and agri (including supply chain), retail CPG, manufacturing, and life sciences, where we continue to build deep domain expertise.
As CTO of Mindsprint, how do you define your core priorities for driving technology transformations, and what key milestones have you achieved so far?
As CTO, my core focus has been on building differentiation for Mindsprint. In the highly commoditised IT services space, we didn’t want to be just another systems integrator. Our goal was to create future-ready, innovative solutions that set us apart.
Interestingly, Mindsprint’s journey began around the same time ChatGPT was launched, which helped shape our strategy to become an AI-first organisation. Over the past 18 months, every major initiative has aligned with this vision. One of our key milestones has been the scale at which we’ve rolled out GenAI and Agentic AI, achievements that, in my view, outpace many others in the industry over the last 18–24 months.
How have you integrated AI into your organisational offerings?
From an organisational standpoint, we’ve built a strong AI-first approach. Our Technology Innovation Group, now rebranded as Future Forward Labs, constantly experiments with emerging tech and works closely with various business practices to integrate AI across all offerings.
While we do offer foundational services like ERP, infrastructure, and data analytics similar to others, our differentiation lies in embedding AI into every layer of delivery. This AI-led focus has been our mantra for the past 12–18 months.
We also believe the traditional IT services model, scaling by headcount, is evolving. In the next 5–7 years, customers will move toward outcome-based engagements. To stay profitable and relevant, we’re shifting from services-led to solution- and IP-led models. Just like OEMs lead with products, we aim to enter customer conversations with pre-built solutions, like in supply chain optimisation, where 60%–70% is already developed, and then customise the rest. This approach ensures faster, more effective implementation and long-term value.
So, with AI enhancing productivity, it’s more about per-unit productivity than simply headcount, correct?
Absolutely. The game is shifting towards efficiency and efficacy. Customers are primarily interested in whether a solution helps them become more efficient, improves quality, or acts as a game-changer by bringing in new revenue streams. These are the three key drivers for digital transformation and non-discretionary investments. IT services organisations must also internalise this: are we building, giving, or working on something that solves any of these three core problems for the customer?
With such rapid technological advancements, how do you balance experimenting with cutting-edge technologies while ensuring a robust security posture and compliance?
That’s a great question, and one I’ve addressed many times over the past 18 months. We take a very balanced approach, our priority is to avoid creating any additional risk for our customers or ourselves. Even a small mistake can have serious consequences, and we can’t afford to compromise the trust our customers place in us.
All experimentation happens in a controlled lab environment, completely isolated from production systems. Every organisation needs such a setup. We’ve established an AI Council and a governance model to evaluate experimental solutions before scaling them to enterprise-grade. It’s not just about running POCs anymore; the focus is on how many of those experiments become robust, production-ready solutions.
At any time, our Future Forward Labs may have 100–150 experiments running. But success is measured by how many of those turn into secure, enterprise-grade deployments. The FFL lead is evaluated based on that outcome.
Security and compliance are tightly integrated into this process. Our AI Council reviews each solution’s security posture, and our in-house InfoSec team continuously runs vulnerability checks. Before every release, even minor ones, we conduct detailed audits and issue a security certification that assures customers their data is safe and won’t leak to the cloud.
Beyond the AI hype, another technology gaining momentum is quantum computing. What sparked your interest in this domain, and what are your planned initiatives in exploring its potential?
In my view, quantum computing is a technology that will evolve gradually, it’s not immediately at our doorstep, but there’s growing momentum. Globally, around $55 billion is being invested annually across 34 countries, including India, with leaders like China, Australia, the US, and parts of the EU pushing significant research.
Quantum computing fundamentally changes how computation works by using qubits instead of bits, allowing for superposition and entanglement. This enables a massive leap in processing power, especially for areas like climate modeling, finance, and logistics.
While we’re still 0–5 years away from stability, due to challenges like qubit sensitivity and error rates, platforms like Google’s Sycamore, AWS Braket, and IBM’s Qiskit are paving the way. Innovation teams are now identifying low-risk use cases that can tolerate minor inaccuracies. For example, it’s not suitable for drug discovery yet, but it can work well for inventory or supply chain predictions.
We’ve already begun experimenting with tools like AWS Braket to build small-scale use cases. The next five years will be about experimentation, and the potential is massive, what takes supercomputers months today could be solved in seconds with quantum computing.
Blockchain technology hasn’t garnered as much attention in India’s tech landscape. What do you think is its potential, and why is it still under-explored here?
Blockchain has had a mixed journey over the past 8–9 years. It often felt ahead of its time, especially in highly regulated industries where concerns around traceability and audit trails cause hesitation.
That said, it has found real use in specific sectors, like retail. For example, some premium perfume brands use blockchain to ensure end-to-end traceability, allowing consumers to scan a code and trace the product’s journey from flower sourcing to final delivery. This helps prevent counterfeiting and builds trust.
However, implementing blockchain is costly, it demands significant infrastructure and maintenance, making it viable mostly for premium products where customers are willing to pay more for authenticity. In commodity-driven industries like ours, where margins are tight and buyers focus on basics like sugar or grains, it’s harder to justify the added cost.
Still, as technologies like quantum computing and AI evolve, they could help reignite interest in blockchain. Just like smaller, AI-powered LLMs have made edge computing more relevant, advances in compute power could eventually make blockchain more accessible and practical across sectors.
Mindsprint recently opened offices in Australia and Singapore. How is India contributing to these new operations?
India remains our core delivery and innovation hub, and we’re fully committed to that. While we do have some talent at customer locations, most of the work is executed remotely from India. For example, after opening our Australia office in April, we’re already serving three clients there, including deploying an AI solution on ServiceNow for a large financial bank, with all delivery handled from India.
We’re also planning to open another office in the Middle East, alongside our existing centre in Dubai. The quality of talent in India is exceptional, and with AI now integrated into B.Tech and M.Tech curricula, we have a strong and growing talent pool. Our focus is on nurturing this talent and continuing to drive global innovation from India. While we may explore innovation centres in other regions in the future, India will remain our primary base for delivery and innovation.
Finally, what’s the vision for Mindsprint as a technology organisation, and how do your tech investments align with broader business goals?
Our vision at Mindsprint is to grow above the industry average by strategically investing as an independent P&L organisation. We’re focused on enhancing our offerings, go-to-market strategies, and customer experience through innovative solutions, IPs, and services, particularly via our Future Forward Labs, where we’re developing differentiating IPs and accelerators.
We’re also building a strong leadership team from progressive tech-driven organisations. Internally, we’ve certified nearly 3,100 out of 3,500 employees in GenAI and are now moving toward persona-based certifications to ensure AI proficiency across all roles, developers, testers, finance, marketing, and sales.
We’ve made significant investments in AI literacy and have built a “tools radar,” a curated set of AI tools, vetted by InfoSec and enterprise architecture, accessible to each function like an internal app store. For example, developers can use tools like Cursor.ai. This AI-first, experimentation-driven approach runs across the entire organisation, not just tech and delivery teams.
You mentioned the roles of developers and testers. With AI implemented in their operations, what positive and negative impacts do you foresee on their roles?
On the positive side, AI brings significant gains in speed and efficiency, what earlier took 10 hours can now be done in 4 to 5. This productivity boost should ultimately benefit the customer, not just allow employees to finish early. For example, if we delivered 50 story points per sprint two years ago, we should now aim for 75 at the same cost. Faster delivery and shorter project timelines will soon become industry expectations, giving early adopters a temporary edge. This also requires a shift in mindset, developers and engineers must evolve to reflect a more progressive, AI-proficient culture.
On the flip side, over-reliance on AI for tasks like code generation or testing could weaken self-thinking and core skills. It’s easy to tell when an email or code is AI-generated, and we don’t want natural intelligence to decline due to dependency on tools. AI should amplify capability, helping people achieve 2x or 3x more, but not replace human understanding.
This is especially important with fresh graduates. While our bootcamps now include AI training, we pair juniors with senior developers for mentoring. We even review code to see how much is AI-generated, as experienced professionals can easily tell. We reinforce that basic programming principles, like object-oriented design, remain critical. The goal is to embrace AI while ensuring developers still master core fundamentals needed to understand, write, and discuss good code.