Agent Orchestrator: 10 Specialized AI Agents Introduced

On March 21, 2025, Adobe announced Agent Orchestrator within Adobe Experience Platform, introducing 10 specialised AI agents designed to orchestrate complex commerce workflows. This represents a significant leap in automation sophistication—moving beyond single-purpose agents to coordinated agent ecosystems.

What is Agent Orchestrator?

Agent Orchestrator is a framework enabling multiple AI agents to work together autonomously on complex tasks. Rather than a single agent handling all requests, the orchestrator delegates tasks to specialists: one agent handles inventory queries, another manages returns, another optimises pricing. These agents communicate, share context, and execute decisions with minimal human intervention.

The 10 Agents

Adobe announced 10 initial agents covering critical commerce areas: customer service (answering questions and escalating issues), inventory management (stock allocation and replenishment), order processing (fulfillment and tracking), returns management (processing refunds and exchanges), pricing optimisation (dynamic pricing based on demand), personalisation (recommending products based on customer history), fraud detection (identifying suspicious transactions), customer loyalty (managing rewards and promotions), content management (updating product information), and analytics (generating business insights from sales data).

Orchestration in Action

Consider a customer ordering a product that becomes out of stock. The orchestrator detects the inventory issue, triggers the inventory agent to check availability at other facilities, coordinates with the logistics agent to arrange alternative shipping, notifies the customer service agent of the situation, and updates the customer autonomously. All this happens without human intervention—the orchestrator makes decisions and executes them.

Enterprise Implications

For large merchants, Agent Orchestrator eliminates manual workflows that slow operations. Returns processing that once took hours becomes automated within minutes. Customer inquiries that required human research can now be handled by agents accessing real-time data. Merchants can redirect human staff toward strategic work rather than routine transactions.

Learning and Adaptation

The agents learn from outcomes—discovering which product recommendations convert, which shipping options reduce returns, which pricing strategies optimise margin. Over time, the orchestrator becomes increasingly sophisticated, making better decisions with less oversight.

Implementation Considerations

Agent Orchestrator requires clear rules defining agent authority—when agents can execute decisions independently versus when they need human approval. Most merchants implement hybrid models: agents handle routine situations autonomously whilst escalating exceptions. This balance maintains operational control whilst capturing automation benefits.

Looking Forward

Agent Orchestrator signals Adobe's vision for autonomous commerce—platforms that anticipate merchant needs and execute decisions in real time. For merchants seeking competitive advantage through operational efficiency, Agent Orchestrator is transformational. For Tom&Co, this opens conversations with merchants about workflow transformation and agent implementation strategy.