Agents in action – Redefining retail ops & marketing

From the Retail Re:Mix panel discussion: “Agents in Action – Redefining Retail Ops & Marketing” September 2025, Dubai

Moderated by Siddharth Gopalkrishnan, COO of Netcore Cloud, this Retail Re:Mix conversation brought together Rehana Raj, Head of Operations – Retail at Choithrams UAE, and Shriram Srinivasan, Director Digital, for a candid look at what happens when AI stops being a silent supporter and starts acting like an active partner.

This wasn’t a conversation about theory. It was about the messy, high-pressure reality of retail, where margins are thin, expectations are rising, and systems that can think and act are no longer “nice to have.”

From silent support to active agents

For years, AI was framed as a back-office tool: chatbots to answer questions, algorithms to crunch data, automation to cut costs. Useful, but limited. The discussion in Dubai made clear that the narrative is shifting.

Agentic AI is no longer waiting for instructions, it’s shaping outcomes. Whether it’s anticipating customer needs, reimagining discovery, or optimising logistics in real time, the new wave of AI isn’t passive. It’s in motion, actively driving both operations and marketing in ways traditional tools never could.

Efficiency meets empathy

Rehana Raj brought the operator’s perspective, reminding the room that grocery and essentials retail leaves no room for error. Efficiency is critical, but so is empathy.

AI can reorder stock, optimise delivery routes, and predict demand with remarkable accuracy. But if it doesn’t keep sight of the customer — their frustrations, their loyalty, their expectations, it fails. Rehana stressed that AI has to be an enabler of trust, not just efficiency. In a market where convenience is non-negotiable, the winning formula is speed with sensitivity.

Context over hype

Shriram Srinivasan pushed the discussion into the digital frontlines. He argued that the retail world is drowning in “AI-for-everything” noise. Too many companies deploy AI because it looks innovative, not because it solves a real problem.

Context, he insisted, is everything. The most impactful use cases are those tied directly to business outcomes: hyper-personalisation that lifts conversion rates, automation that reduces customer friction, predictive analytics that cut waste. Everything else? Expensive theatre. The message was blunt: AI without context isn’t innovation — it’s distraction.

Guardrails, risk, and reality

Moderator Siddharth Gopalkrishnan steered the group toward the inevitable risk conversation. And here, the candour was striking. The panel agreed that AI adoption will fail as often as it succeeds. Experiments will flop, pilots won’t scale, customers will push back. That isn’t weakness, it’s part of progress.

But risk needs boundaries. Without guardrails around data, ethics, and transparency, the cost of failure escalates from wasted budget to broken trust. The retailers who get this right will treat AI like a disciplined portfolio: test aggressively, measure tightly, kill quickly when things don’t work, and double down where value is clear.

Already here, not “someday”

Perhaps the most powerful takeaway was that agentic AI isn’t some distant future. It’s here, now, quietly rewriting the rules of retail. The question is no longer whether to adopt, but how to adopt responsibly, and how fast.

The consensus was clear: waiting on the sidelines is no longer safe. The leaders will be the ones who accept that experimentation is survival, that risk is the strategy, and that courage will separate the retailers who thrive from those who fade.

The bottom line

Retail can’t afford to think of AI as just a helper anymore. The shift to agents in action means redefining how operations run, how marketing connects, and how businesses are built. For Rehana Raj and Shriram Srinivasan, the challenge isn’t whether AI belongs in retail — it’s how to use it with purpose, empathy, and accountability.

Because in retail today, safe doesn’t scale. Agents do.