The Internet Was Built for Humans. AI Is About to Change That.

Mar 5, 2026

Maju Kuruvilla, Founder and CEO of Spangle, builds commerce infrastructure for an agent-driven future. She explores how AI buyer agents will change shopping, why holding context beats relying on identity, and how product-level intelligence can bridge online and physical retail. The conversation covers agent versus human UX, ambiguous requests like “buy a red sweater,” and privacy-friendly ways to keep context across channels.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

INSIGHT

Context Is More Valuable Than Identity

Context matters more than customer identity for conversion because ad-driven visits often lose the original inspiration when sent to generic pages.
Maju explains nearly half of commerce traffic originates elsewhere and websites commonly restart journeys, killing the signal agents or humans need.

INSIGHT

Seller Agents Must Detect And Serve Buyer Agents

Commerce will split into buyer agents and seller agents that must detect modality (agent vs human) and respond differently.
Spangle builds a seller-agent intelligence layer that curates merchant responses, schemas, discounts and UX per incoming modality.

INSIGHT

Match Intent By Expanding Both Consumer And Product Contexts

Matching intent requires expanding context on both sides: consumer signals and enriched product descriptions.
Maju describes using clicks, ad provenance and product uses (events, occasions, shades) to embed in a vector space for matching.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

What happens to e commerce when AI agents start shopping instead of humans?

Maju Kuruvilla, Founder and CEO of Spangle, joins the show to unpack a shift most companies are not prepared for. If AI agents become buyers, the entire digital shopping experience must change. Websites today are designed for human psychology, not machines making decisions.

In this conversation, Maju explains why context is becoming the most important layer in commerce. From marketing clicks to storefront visits, most companies lose the context that originally inspired a purchase. The future belongs to systems that can capture, carry, and act on that context across every channel.

The discussion explores agent driven shopping, the limits of traditional customer data systems, and how AI can reshape both online and physical retail experiences.

Key Takeaways

• Context matters more than identity. Knowing what someone is trying to do right now is often more valuable than knowing who they are.

• Most e commerce experiences reset the customer journey. When someone clicks from an ad to a site, the original inspiration is usually lost.

• AI agents will shop differently than humans. They are not influenced by visual design or marketing psychology the same way people are.

• Commerce will not become fully agent driven. Instead, brands must design experiences that work for humans, agents, and hybrid interactions.

• Physical retail may benefit the most from AI driven context because stores can blend digital signals with real world behavior.

Timestamped Highlights

00:00 Why the next generation of e commerce will be built for AI agents, not just human shoppers.

02:08 The hidden problem in online shopping today. Most websites lose the context that brought the customer there.

06:11 Buyer agents and seller agents. How commerce may evolve into AI systems negotiating purchases.

11:38 Why a simple request like “buy a red sweater” is actually a complex problem of interpretation and context.

16:30 How AI could transform physical stores through dynamic recommendations and real time shopping guidance.

22:30 Why collecting endless customer data might be the wrong approach to personalization.

27:59 The future of autonomous shopping and why personal AI agents may eventually handle everyday purchases.

A Moment That Sticks

“Context is what matters. The fact that I bought a TV before is interesting, but not important. What matters is what I am trying to do right now.”

Practical Insight for Builders

If you are building AI driven commerce tools, start with the product layer.

According to Maju, the foundation is making your product catalog intelligent. AI systems need rich product understanding so they can match intent with inventory. Once the catalog becomes machine readable and context aware, everything else becomes easier to automate.

Call to Action

If you enjoyed this conversation, follow the show and share this episode with someone working at the intersection of AI, commerce, or product development.

New conversations every week with the builders shaping the future of technology.