From orchestration to outcomes: New agentic workflows and domain agents in IBM watsonx Orchestrate

Organizations across industries are moving beyond experimentation and into the era of agentic AI—where intelligent agents don’t just respond to requests, but reason, collaborate and act to deliver real outcomes.

Today, we’re excited to announce new capabilities in IBM watsonx Orchestrate that take agentic AI to the next level:

• Agentic Workflows enables developers to build standardized, reusable flows that sequence multiple agents and tools consistently and reliably, so they no longer have to rely on brittle scripts that break when scaled.
• Langflow integration with watsonx Orchestrate enables developers to continue building agents and flows with a drag-and-drop visual builder.
• Agent Governance and Observability transforms agent evaluation into a full observability experience, from pre-deployment through production.
• To add to our existing catalog of pre-built agents in HR, Procurement and Sales, we are releasing new pre-built agents for Finance and Supply Chain to bring deep industry expertise and accelerators to common business challenges.
• Pre-built Customer Service agents help customers resolve their issues quickly and accurately through the phone, chat or text.

These innovations empower enterprises to move faster, with confidence, by orchestrating intelligent agents that are both adaptive and accountable.

Orchestrate is designed to be tool-agnostic and adaptable to any environment, enabling scalable deployment and governance of AI agents. Traditional monitoring tools often fall short when it comes to tracking AI agents. These agents not only require monitoring but also necessitate a feedback loop to learn and improve continually. The distributed nature of modern AI systems further complicates matters, making distributed tracing essential.

New AgentOps capabilities in watsonx Orchestrate include a built-in observability and governance layer that provides full lifecycle transparency. With real-time monitoring and policy-based controls, AgentOps ensures agents behave reliably and securely.

• Agent Observability empowers agent builders with a holistic view of their agent environment. They can effortlessly monitor usage, success rates and latencies, all from a unified dashboard.
• Agent Governance capabilities in watsonx Orchestrate provide a robust framework for pre-deployment evaluation of agents. Using the Agent Development Kit (ADK) or the no-code Agent Builder, developers can evaluate their agents across multiple dimensions including journey completion, answer relevancy and tool call accuracy.
• Orchestrate will also support Production Monitoring of AI agents to ensure that predefined guardrails and policies are enforced automatically, preventing issues like prompt injection attacks and unauthorized data access.

AI agents are designed to act autonomously. But when accuracy, compliance and repeatability are critical, autonomy needs structure. That’s where agentic workflows in watsonx Orchestrate come in. And with the integration of Langflow, users can design, visualize and manage complex flows that combine:

• Predefined toolchains – ensuring the right systems are used in the right sequence
• Conditional logic – guiding agents with branching, retries and checkpoints
• Data handling transparency – controlling how inputs and outputs move through each step.

The result: workflows that are predictable, auditable and reusable—giving organizations confidence that critical processes will run correctly every time.

From financial approvals and compliance checks, to supply chain exception handling, to customer case routing, these structured flows ensure AI agents deliver business-grade reliability.

Alongside workflows, we’re introducing the next set of watsonx Domain Agents—pre-built multi-agent systems designed to address some of the most complex and high-value enterprise functions.

Powered by IBM Planning Analytics, Finance Agents accelerate planning, forecasting and scenario analysis while automating repetitive tasks like budget allocation, reconciliations and risk assessments. CFOs gain faster insights, speed up planning cycles and acquire more time to focus on strategy.

Designed to improve agility and resilience, Supply Chain Agents help accelerate response to disruptions, optimize inventory and streamline order management. With built-in integrations to IBM Sterling, SAP, Oracle, Coupa and more, these agents deliver measurable efficiency and cost savings.

Customer Service leaders can now deploy AI agents to reduce case resolution times, improve customer experience and scale support without increasing costs. Agents handle routine inquiries, surface knowledge for human contact center works and automate back-end workflows—freeing human teams to focus on higher-value interactions.

• For Finance leaders: 79% of CFOs say they are accelerating AI transformation, but reliability and governance remain barriers. Orchestrate brings responsible, transparent automation to the finance function.
• For Supply Chain leaders: 62% report agentic AI accelerates speed to action, and 76% see improved process efficiency. Orchestrate provides resilience and agility at scale.
• For Customer Service leaders: With customer expectations at an all-time high, AI-driven care agents help organizations deliver faster, more personalized service while reducing operational costs.

This launch builds on IBM’s vision of putting AI to work across the enterprise—combining reasoning, orchestration and execution in a unified platform.

With agentic workflows and domain agents now in Orchestrate, enterprises have the tools to:

• Scale AI faster, with pre-built agents and flows
• Customize to their business using no-code and pro-code tools
• Govern agentic systems with enterprise-grade transparency and security

See watsonx Orchestrate in action