August 5, 2025
5 min read
Patrick Kulp
Walmart consolidates its AI agents into four ‘super agents’ to streamline tasks for shoppers, employees, suppliers, and developers.
Innumerable AI agents can overwhelm. Can Walmart’s ‘super agents’ come to the rescue?
For the last few months, Walmart has been rolling out an army of AI agents to assist its workers, according to global chief technology and development officer Suresh Kumar. These AI systems help with tasks such as pricing items, stocking shelves, or requesting time off. However, managing a multitude of individual agents has become overwhelming for users.“If you build individual agents for every use case, it very quickly becomes overwhelming for the end user,” Kumar explained.To address this, Walmart is reorganizing its many agents under four “super agents.” These advanced personas can route queries to any of the dozens of sub-agents within their scope, simplifying interactions. The four super agents focus on distinct groups:
- Shoppers
- Store associates
- Suppliers and advertisers
- Developers Currently live is Sparky, a shopping assistant that helps plan around occasions, summarizes reviews, and sets up recurring household orders. Another super agent, Marty, assists merchants with crafting ad campaigns and listing items. The internal super agents for employees and developers remain unnamed.
- Walmart Bets on AI Super Agents to Boost E-commerce Growth
- AI Driven Crypto Trading Tools Reshape Market Strategies in 2025
- Understanding AI Agents: Capabilities, Risks, and Growing Role
“The agent knows that you are a merchant. The agent knows that you are a store manager. The agent knows that you work in people services. And it also knows all our sub-agents that are there. It can coordinate,” Kumar said. “We will have tons and tons of sub-agents underneath as we start automating more capabilities, as we start addressing more use cases.”
The future of AI agents: teamwork and orchestration
Industry experts have long predicted that AI agents will evolve from isolated tools into orchestrated networks of task-specific entities. Babak Hodjat, CTO of AI at Cognizant, noted in a previous interview:“The dominant discourse in modern LLM-based agents is still leaning heavily toward building large single agents as one-stop shops, and I don’t think that scales. Multi-agent systems hold the key to leveraging AI in ways that can balance and analyze often conflicting needs to deliver higher value impact.”
Trust and governance challenges
Despite the promise, most companies are still early in their AI agent adoption. A recent EY survey found only 14% of companies have fully implemented agents. The biggest barriers are trust-related, including cybersecurity and data privacy concerns. Walmart has established a framework to govern access, security, and compliance. Typically, a human remains in the loop to evaluate responses and fine-tune models when outputs are misaligned.“[We] realize that this is a journey, just like all of these other businesses,” Kumar said. “So it’s not like we are going to start on day one giving full access and full delegate authority and full capabilities to any single agent, even the super agent.”
Interoperability and the shopping experience of the future
Walmart is building its agents to coordinate with data sources through Anthropic’s Model Context Protocol (MCP), an open-source framework that standardizes how agents connect with external systems. The goal is full interoperability, allowing agents like Sparky to continue shopping journeys started on other platforms, or for store associates’ super agents to communicate with external AI agents, such as those from Aetna for health benefits. Kumar envisions a future where shopping experiences span multiple media formats, including TikTok videos and photos of home decor.“If I’ve already decided on a particular decor for my living room, what is the particular painting that I want to hang on the wall? This is not something that you can…do through search keywords,” he said. “No, you actually take a picture and you let the agent go, ‘Ah, now I understand what it is that you are trying to do: Here, let me suggest a bunch of stuff.’ So I think that is the future.”