Supply chain software has spent years adding predictive analytics and AI features to planning systems, but ketteQ is arguing that the next step is fundamentally different: letting users ask an AI system questions that were never programmed into a workflow. The supply chain orchestration company has announced general availability of Quintus, an agentic AI platform designed to reason over live supply chain data, generate and execute Python code, test thousands of planning scenarios and operate above existing ERP and planning systems.
Supply chain leaders rarely have the luxury of asking only the questions their software was designed to answer.
When a supplier shuts down unexpectedly, a planner may need to know which customer commitments are exposed, what inventory can be redirected, how revenue could be affected and which alternative scenarios minimize the disruption. Traditional planning systems can answer many of those questions, but typically through predefined workflows, scheduled runs or previously calculated scenarios.
ketteQ wants to change that interaction model.
The supply chain orchestration company has announced the general availability of Quintus, an agentic AI system it describes as “Free-Range AI” for supply chain operations. The platform is designed to reason over questions and tasks dynamically, learn skills and execute missions within a governed and auditable framework.
Rather than replacing an enterprise resource planning or supply chain planning platform, Quintus is designed to operate above existing systems. ketteQ says it can sit over platforms including SAP IBP, Kinaxis, o9 and Blue Yonder, or operate alongside ketteQ’s own planning and execution products.
That positioning is significant because supply chain software is unusually fragmented. Large manufacturers and distributors often have years of investment in ERP, planning, warehouse, transportation and customer systems. A new AI capability that requires replacing the underlying architecture faces a much higher adoption barrier than one that can consume data from the existing stack.
Quintus is attempting to occupy the decision layer above that infrastructure.
The company’s most distinctive technical claim centers on how the system solves problems. ketteQ says Quintus can generate and execute Python code dynamically and uses PolymatiQ, its patent-pending agentic solver, to evaluate thousands of constrained planning scenarios simultaneously.
That is different from simply querying a database or asking a generative AI model to summarize a report.
Consider a supplier shutdown. A conventional dashboard might show current inventory and open orders. A planning engine could run a disruption scenario during its next planning cycle. Quintus is intended to reason over the situation dynamically and identify the consequences and possible responses using current data.
The distinction is between reporting what has already been calculated and calculating an answer to a new question.
That model is increasingly familiar in software development and knowledge work. ChatGPT and Claude demonstrated that users can interact directly with general-purpose reasoning systems rather than navigating menus designed around predefined workflows.
ketteQ is applying the same interaction principle to supply chain operations.
A planner could ask what happens if a Vietnamese supplier’s plant goes offline for two weeks. A CFO could ask how the disruption affects revenue at risk. A salesperson could ask whether a customer’s delivery commitment remains achievable.
The interface is also intended to extend beyond traditional planning software. ketteQ says users can interact with Quintus through channels including email, Microsoft Teams and Slack.
That could make supply chain intelligence less of a specialist function. If the system works as intended, commercial, finance and operations teams can ask questions against the same underlying data without becoming experts in a planning application.
The architecture is also where ketteQ is differentiating its product from established planning vendors.
The company argues that many incumbent supply chain platforms were built around older architectures and are now adding AI as an additional layer. ketteQ, by contrast, says its platform was designed as an open architecture beginning in 2019, allowing Quintus to function as a neutral decisioning layer.
That is a vendor claim rather than an independently established industry conclusion, but the underlying competitive issue is real. Vendors such as SAP, Kinaxis, Blue Yonder and o9 have substantial installed bases and increasingly offer their own AI capabilities. Their advantage is deep integration with their respective platforms; ketteQ is betting that customers will value an AI layer that can reason across multiple systems.
The company says Quintus can be deployed over existing platforms in four to eight weeks, avoiding a rip-and-replace project. Enterprise buyers should treat that as a deployment claim to validate during procurement rather than a universal implementation benchmark.
ketteQ also says the platform is already running in production at organizations including Alliance Consumer Group, JCI, Mativ, NCR and Zeus.
At Alliance Consumer Group, the company reports that inventory turns improved from 2.75 to 4.0 and that real-time available-to-promise answers can be delivered through Salesforce in under 15 seconds during sales calls. Those are company-reported results; independent verification would be needed to determine how much of the improvement is attributable specifically to Quintus.
The bigger question is governance.
Supply chain decisions can affect inventory, working capital, customer commitments, production schedules and revenue. An AI system that can reason dynamically therefore needs more than an impressive conversational interface.
ketteQ says every Quintus decision is auditable, explainable and subject to human override. That is an important requirement if AI moves from analysis into execution.
The emerging enterprise AI market is increasingly separating copilots that recommend from agents that act. The latter can create much greater productivity gains, but they also require stronger controls around authorization, traceability and failure recovery.
Quintus is positioned somewhere between the two: it can reason and act on supply chain tasks, but within a framework ketteQ says remains governed by humans.
For supply chain organizations, that could be the more consequential part of the launch.
The industry’s AI race is not simply about adding a chatbot to planning software. It is moving toward systems capable of understanding a changing operational environment, testing alternatives and helping teams decide what to do next.
If ketteQ can deliver that capability across heterogeneous enterprise systems without requiring customers to replace their existing planning infrastructure, Quintus could represent a different approach to supply chain AI: not another feature inside the planning suite, but an intelligence layer sitting above it.
The market will ultimately judge that proposition on a harder metric than conversational fluency — whether the system consistently makes better decisions, quickly enough to matter, and safely enough to trust.
Market Landscape
Supply chain planning is becoming a major enterprise AI battleground.
Established platforms such as SAP, Kinaxis, Blue Yonder and o9 Solutions are incorporating AI, automation and predictive capabilities into planning environments. Meanwhile, cloud platforms from Microsoft, Amazon and Google are providing the infrastructure and AI services enterprises can use to build their own supply chain applications.
The competitive distinction is increasingly architectural.
Traditional planning systems often depend on structured planning cycles, optimization engines and predefined workflows. Agentic AI introduces the possibility of dynamic reasoning across data sources and business constraints, potentially allowing users to pose new questions without waiting for a new workflow to be configured.
McKinsey has estimated that generative AI could create significant value across supply chain and inventory-management activities, particularly through improved forecasting, inventory optimization and decision support. The consultancy has also emphasized that capturing that value requires changes to operating models, data foundations and workforce capabilities.
For enterprise buyers, the central question is therefore not whether a vendor offers “AI.” It is whether the AI can work across existing systems, use current data, produce auditable decisions and integrate into the people and processes responsible for supply chain outcomes.
ketteQ’s Quintus launch is aimed squarely at that gap.
Top Insights
ketteQ has made Quintus generally available as an agentic AI layer designed to reason over live supply chain data and existing planning systems.
Quintus uses dynamically generated Python and ketteQ’s PolymatiQ solver to evaluate thousands of constrained scenarios rather than relying solely on precomputed planning outputs.
The platform is designed to sit above systems including SAP IBP, Kinaxis, o9 and Blue Yonder, potentially reducing the need for rip-and-replace transformations.
ketteQ reports production deployments at multiple companies, while Alliance Consumer Group has reported improved inventory turns and faster available-to-promise answers.
Human override, auditability and explainability are central to ketteQ’s positioning as agentic AI moves from recommendations toward operational decision-making.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
