Blue Yonder Bets $2.5 Billion on AI Agents to Replace Supply Chain Apps

Quick Facts

• Blue Yonder spent $2.5 billion over four years rebuilding its technology stack around AI agents, posting $1.42 billion in revenue for FY25.
• The company deploys named, governed AI agents with persistent memory built on a unified data model powered by the Snowflake Data Cloud.
• A new Model Training Factory, built on NVIDIA Nemotron, trains specialized agents on synthetic data to run supply chain decisions autonomously.

Blue Yonder Inc. is redesigning its supply chain platform around autonomous AI agents, moving away from application-centric software toward what CEO Duncan Angove calls a permanent state of operational disruption management. The Scottsdale, Arizona-based company serves more than 3,000 retailers, manufacturers, and logistics providers and generates over 20 billion predictions daily.

Angove made the case directly: “The old supply chains were built around forecasting demand and then orchestrating an entire value chain around that. And that just doesn’t work in today’s world.”

The company has spent four years and $2.5 billion rebuilding its technology stack from the ground up. In FY25, Blue Yonder reported $1.42 billion in total revenue, 10.4% SaaS revenue growth year-over-year, net revenue retention of 103.8%, and 114 new customers.

Agents Replace Applications

Blue Yonder’s architecture centers on a concept its leadership describes as “the agent is the app.” Rather than layering AI onto existing interfaces, the company is stripping out human confirmation steps that agents no longer require.

Chief Design Officer Nunzio Esposito described the approach: “We don’t believe an agentic experience is just chat. The way that we’re looking at it is ensuring that it is infused into the fabric of the experience. It’s not a sidecar or bolt-on.”

The agents are not stateless assistants. Blue Yonder deploys named, governed digital workers with persistent memory that operate natively on the Blue Yonder Platform and share a single unified data foundation powered by the Snowflake Data Cloud. That shared model allows the Network Agent to communicate with the Warehouse Agent directly, without relying on brittle API connections.

Angove framed the human role shift plainly: “You have to design for two participants.” Planners move from executing tasks to setting objectives, while agents handle real-time signal ingestion, multi-variable trade-off modeling, and decision execution against systems of record.

New Agents and Platform Updates

In March 2026, Blue Yonder expanded its agent lineup with a Fulfillment and Sourcing Agent in beta, new manufacturing planning agents that automate issue detection across demand, supply, and inventory plans, and expanded AI functions for transportation management covering load monitoring, weather-linked alerts, and backhaul identification.

Chris Burchett, SVP of Generative AI at Blue Yonder, described the current state: “For the first time in the industry’s history, we have end-to-end supply chain operations and solutions built on one common data model. Our agents are a huge part of that now, with the warehouse ops agent, logistics ops agent and inventory ops agent.”

The company also released an Orchestrator mobile application for on-the-go access to agentic AI and added Microsoft Teams integration to bring AI-driven actions into existing collaboration workflows.

NVIDIA Partnership Adds Specialized Training

At its ICON conference in May, Blue Yonder announced a Model Training Factory built on NVIDIA Nemotron. The factory produces specialized AI agents trained to perform high-value supply chain tasks at the level of subject matter experts, covering warehouse management, supply and demand planning, transportation, merchandising, and network operations.

The models are trained on synthetic data, a detail that matters for enterprise customers concerned about proprietary data exposure during model development. Agents produced by the factory are fine-tuned for complex, multi-step workflows and graded against quality benchmarks before deployment.

The Blue Yonder 2026.2 platform release packages these capabilities under what the company calls the Agentic Supply Chain, a unified environment for AI, planning, execution, and network operations on a common platform. The target state is autonomous supply chain operations where agents continuously monitor the network and execute decisions while humans manage strategy and exceptions.

Read more: Blue Yonder redesigns supply chain operations around AI agents