AI is only as strong as its data foundation: Sean Stauth, Global Director – AI & ML, Qlik
In an exclusive interaction with Express Computer, Sean Stauth, Global Director – AI & ML at Qlik, delves into how agentic AI is redefining enterprise value creation beyond chatbots. As Qlik strengthens its India strategy, Stauth discusses embedding AI into operational workflows, the challenges of building responsible multi-agent systems, and how data readiness is becoming the bedrock for sustainable AI. With actionable insights from manufacturing to healthcare, Qlik aims to empower Indian enterprises to not only adopt AI but scale it ethically, securely, and profitably.
AI conversations often begin with chatbots, but Qlik’s work reveals a far wider canvas. Why do you believe the future of AI lies beyond chatbots, and what new possibilities does agentic AI unlock for enterprises?
That’s an excellent starting point. If we trace AI’s evolution, we saw an inflection in 2010 when large-scale compute enabled widespread enterprise adoption of predictive analytics. But generative AI comes from a different research lineage—rooted in natural language processing, with the transformer architecture as the breakthrough. It allowed for the creation of large language models (LLMs) and the agentic systems we see now.
What excites business leaders today is the natural language interface to data. Executives can now query systems conversationally to understand performance and act on it. These agentic systems are not just chatbots, they’re dynamic, business-specific tools, often combining several models to automate, reason, and optimise processes.
With autonomy comes complexity. What are the key challenges in aligning agentic systems with real-world business goals while ensuring ethical and responsible AI?
We’re seeing the rise of complex multi-agent systems that serve specific business functions. These systems frequently involve orchestration across multiple foundational and small language models, requiring a nuanced balance between autonomy and oversight. One of the primary challenges is ensuring that these agentic systems are tightly aligned with the strategic goals of the enterprise, be it improving operational efficiency, reducing risk, or enhancing productivity. Without a clear objective, there is a risk of misalignment and wasted resources. At the same time, technical complexity must be managed with care, given that these AI stacks are fundamentally different from traditional IT systems. Transparency, modularity, and scalability become non-negotiable when models are making semi-autonomous decisions. Governance is another pressing concern, especially in regions like Asia where there is heightened sensitivity to AI sovereignty and compliance. Enterprises want to know where their data resides, how it flows, and how outputs are derived. This calls for robust traceability and explainability at every step. Finally, ensuring enterprise-grade readiness in terms of security, access controls, and regulatory compliance is vital for AI systems to scale responsibly. These aren’t just technical considerations—they form the foundation for building trust in AI-powered decision-making across the organisation.
AI is no longer just an innovation layer it’s embedding itself across functions. What real-world transformations have you seen where AI delivered measurable value?
The key difference now is AI’s dynamic nature. Unlike traditional systems, agentic AI can continuously adapt. One compelling example is a manufacturer producing perishable goods. Using Qlik, they predicted monthly volumes with over 90% accuracy, dramatically reducing waste and lost revenue.
In healthcare, Appalachian Regional Healthcare in the U.S. embedded AI into scheduling workflows. The system flagged likely no-shows and recommended interventions, saving over $6 million annually while improving patient outcomes. These aren’t pilot projects—they are ROI-generating, life-impacting use cases.
Indian enterprises often cite data readiness as a challenge, especially with legacy systems and evolving compliance norms. What’s your advice for building a secure, scalable data foundation tailored to Indian needs?
That’s a critical point. AI is only as strong as its data foundation. In India, many enterprises still rely heavily on legacy systems like SAP. Our platform helps them unify data structured, unstructured, on-premises, or cloud into trusted, curated data products ready for AI consumption.
The focus must be on traceable, curated data pipelines. Feeding raw or unverified data into AI risks producing flawed insights. We offer deep integration capabilities that make Qlik a bridge between legacy and AI-ready infrastructure, essential for Tier-2 and Tier-3 city enterprises aiming to scale responsibly.
Tell us more about Qlik’s embedded AI innovations like the Agentic AI Experience and Discovery Agent. How are these transforming user experience, especially for non-data scientists in India?
Our goal is to reduce time to market. We offer end-to-end tools, from data pipeline creation with Qlik Data Flow to predictive modelling with Qlik Predict. These are designed with a no-code approach so that even non-experts can create agentic AI workflows.
We serve two user profiles: sophisticated data scientists who appreciate rapid prototyping, and business users who can now build models without deep technical knowledge. This approach empowers self-service AI and helps organisations—from healthcare to e-commerce, move from insights to action in high-stakes environments.
With AI moving from experimentation to strategic implementation, how is Qlik evolving to support responsible, transparent AI at scale—especially for Indian enterprises?
We’re observing a clear shift. The big trend now is the rise of agent development frameworks, intra-agent protocols like A2A and MCP, and SDKs that enable custom AI solutions.
Qlik’s focus is on simplifying this complexity with a governed, secure architecture. We’ve built end-to-end data lineage, explainability, and downstream impact analysis into our product. Enterprises can now see not just where insights came from, but how AI decisions were made and what applications use those outputs. This level of transparency is vital for responsible, compliant scale.
Which sectors are seeing the highest traction for AI adoption through Qlik in India and beyond?
Manufacturing is thriving, it’s a data-rich sector where AI improves yield and predictability. In India, we see insurance rapidly embracing AI, especially for claims management, fraud detection, and customer risk profiling. The complexity of the Indian insurance landscape—with its diverse distributors, consumers, and regulatory layers—makes agentic systems even more relevant.
Financial services and supply chain are other high-growth areas. But broadly, any sector with operational complexity stands to gain from agentic AI.
India is also a deep well of tech talent. How is Qlik leveraging this for both market expansion and innovation?
The quality of India’s tech talent is world-class. From Bengaluru to Mumbai, we’ve been meeting partners and customers who are pushing boundaries in AI system design. Our Bangalore office is a key hub. We see India not only as a strategic market but also as a centre for co-innovation.
Our partners, both large SIs and regional specialists, are essential. They deeply understand local industries and help us tailor solutions. We’re actively engaging them in joint go-to-market strategies, capability building, and co-innovation dialogues.
As universities begin embedding AI in their curricula, how do you view the next wave of talent entering this space?
It’s incredibly promising. Today’s graduates already come equipped with Python, data science skills, and familiarity with building agentic systems. Whether they’re from technical or non-technical backgrounds, the mindset is shifting. At Qlik, we strive to build tools that align with their vision and skillsets, so they can hit the ground running.
What’s your advice to Indian enterprises beginning their journey with agentic AI? How can they ensure long-term value over the next 12 months?
I often compare it to the dawn of the internet. Many wonder, “Should we wait for AI to mature?” But this technology doesn’t stand still. It evolves monthly. Waiting risks irrelevance.
Even if 60–80% of AI projects fail, that’s a signal that organisations are experimenting, learning, and iterating. My advice: build your skills, take calculated risks, and treat failure as part of progress. Without experimentation, there are no use cases. The organisations willing to fail early are the ones that will lead tomorrow.
It’s encouraging to see India’s momentum in AI. We’re excited to continue our work in this dynamic market with a strong partner ecosystem and world-class talent.
