For Kumaran, the path to rapid, responsible AI innovation begins with the right infrastructure. Rather than slowing experimentation to protect quality, his approach at CreatorIQ embeds safety mechanisms and guardrails directly into tooling and processes — allowing teams to iterate fast without compromising enterprise-grade reliability.

Central to this is a rethink of ML pipeline architecture. Kumaran emphasises that pipelines must support the deployment and tuning of multiple models simultaneously, while enabling rapid extraction of signals from experimentation and broad feature engineering across capability segments. The ability to test across controlled user segments — both for experiential and non-experiential elements — is a priority that shapes how the technology team operates.

One of the most significant technical challenges in creator marketing is data fragmentation. Instagram, TikTok, YouTube, and other major platforms each structure data differently — creating complexity that compounds at global scale. Kumaran describes how CreatorIQ has addressed this through sophisticated at-scale pipelines and a centralised intelligence layer: the Creator Graph™.

Processing 250 million posts daily, the Creator Graph standardises how performance, suitability, and engagement are measured — enabling brands to compare results across platforms in a consistent, meaningful way. It is, in essence, a unified intelligence foundation for the entire product suite.

On the question of latency, cost, and scalability trade-offs, Kumaran is direct: every major feature must be evaluated against all three dimensions. With AI capabilities, a new set of financial models now applies — encompassing choice of models, token pricing, and decisions on whether to build custom pipelines. But as he notes, the organisational muscle for navigating these trade-offs already exists; AI simply adds new parameters to a familiar framework.

The heart of Kumaran's philosophy is a firm belief that creativity is not easily replicable by machines. While AI at CreatorIQ focuses on understanding patterns in data — not generating content — the creative work always remains human. Complex ML models and pipelines analyse creator, content, and performance data at scale; but strategic, creative decision-making belongs to people.

This principle is embedded directly into product design. Human-in-the-loop AI systems at CreatorIQ are built so that marketers can review, adjust, and override AI-driven recommendations through clear workflows and rapid feedback loops. The system surfaces insights and intelligence drawn from large-scale data — but humans decide what is right for the brand, and control when and how that intelligence is used.

This balance becomes even more critical as brands shift from one-off campaigns to long-term creator partnerships, where trust, authenticity, and relationship depth cannot be automated away.

For enterprises adopting AI at scale, Kumaran identifies flexibility and workflow control as the essential enablers. Rather than demanding organisations adopt new systems wholesale, CreatorIQ's platform is built to progressively meet customers where they are in their AI journey — supporting different internal structures and workflow models at enterprise scale.

On interoperability, the focus is on integrating with the systems organisations already use, creating consistency across workflows and measurements. When payment data is connected across 80+ markets and 60+ currencies, the result is a broad, unified view of ROI — without forcing teams to change how they operate. The goal, as Kumaran puts it plainly, is to reduce fragmentation without forcing disruption.

Stepping into the CTO role at a pivotal moment, Kumaran describes what excites him most as the opportunity to build on a strong existing foundation: deep data assets, a mature platform, and a clear pathway to more advanced, agentic AI capabilities. The company has already begun integrating large language models and custom ML models into its products — with SafeIQ cited as a leading example of how that roadmap is coming to life.

His approach to balancing continuity with transformation is deliberately measured. Two commitments remain non-negotiable: customer delivery and feature quality. Within that, the areas earmarked for disruption include the ability to run experiments at scale, automation and workflow rethinking, and the underlying infrastructure needed to handle both complex pipelines and enterprise-wide workflows simultaneously.

Looking three to five years ahead, Kumaran envisions creator discovery evolving from keyword-based search toward complex, conversational AI interfaces with retained context — intelligent systems that curate creators based on defined outcomes, drawing on the Creator Graph, semantic ML intelligence, and vast historical data.

When asked whether the next phase of the creator economy will be won by proprietary data, AI capability, or ecosystem partnerships, Kumaran's answer is unequivocal: it's the combination of all three. Data is the foundation. AI makes sense of it. Partnerships allow activation across platforms. What will ultimately matter is how well organisations bring those elements together into repeatable, sustainable systems.

The next phase, he argues, is not simply about scale — it is about sustainability. The creator economy only works long-term if value is more predictable and better shared across creators and brands alike. That requires systems designed not just for performance, but for fairness, traceability, and lasting trust — values that are hardwired into how CreatorIQ is building for the future.

Key Takeaways

• →Successful AI adoption demands strong data foundations, thoughtful system design, and human control — not just technical capability
• →The Creator Graph™ processes 250 million posts daily, standardising performance and engagement measurement across all major social platforms
• →Human-in-the-loop systems empower marketers to review, adjust, and override AI recommendations — keeping humans in control of brand decisions while AI handles scale
• →Audit trails, configurable policies, and brand-specific controls are built into the platform to ensure AI decisions are traceable and adjustable at enterprise scale
• →The future of creator discovery will shift from keyword search to conversational AI with retained context, curating creators based on defined outcomes
• →Winners in the AI era will combine proprietary data, AI capability, and ecosystem partnerships — and build them into repeatable, sustainable systems rather than one-off capabilities