Nintex unveils AI tools for governed business workflows

Nintex has launched two tools for building and managing AI agents inside business workflows, as organisations seek to put AI into production without losing oversight or compliance control.

The new tools, Nintex Agent Designer and Nintex Orchestration, are available natively in Nintex CE to a select group of customers. The release targets organisations testing agent-based automation but struggling to connect it to day-to-day processes and governance.

Businesses have moved quickly to evaluate AI agents that can interpret information and take actions. Many teams have found, however, that standalone deployments can create new risks when they sit outside established process controls, approval steps and audit trails.

Nintex positions the releases as a way to bring AI agents and rules-based workflow logic into the same process framework, keeping human decision points visible and applying structured controls where needed.

Blended execution

In practice, the tools aim to combine two styles of automation: deterministic rules and structured workflows, and non-deterministic agent behaviour that applies judgement based on context and retrieved information.

IDC Research Vice President, Software Development Arnal Dayaratna said buyer expectations are shifting. "Rather than replacing structured processes with fully autonomous systems, organisations today want platforms that support both deterministic and agentic approaches across a single orchestration framework," he said.

He added that the blended model supports governance and operational requirements. "A blended approach allows organisations to apply agentic AI where judgment and interpretation add value, while maintaining conditional workflow control where precision and compliance are required. This ability to coordinate both models within business workflows is a critical requirement for scaling AI adoption," he said.

What changes

Nintex Orchestration introduces a phase-based model for process execution, breaking large end-to-end processes into modular phases. Workflows can move between phases dynamically based on context, including repeating or escalating a phase, or rerouting work.

Agent Designer is positioned as the way to embed AI agents into those phases, keeping agents, people and core systems coordinated within the same operational flow. It also provides a single place to design how work moves between automation and human review.

Nintex describes the combined model as a way to apply AI where interpretation and judgement add value while retaining conditional control for higher-risk steps. It also highlights the need for visibility across long-running processes that may span days or weeks and move between teams, systems and decision points.

Agent patterns

Agent Designer supports adaptive AI agents and includes supervisor and multi-agent patterns. Nintex says the agents can plan and execute multi-step actions, retrieve contextual information to inform decisions, and escalate to humans when deterministic control is required.

Orchestration also supports automation components such as RPA, document processing and system connectors. This aligns with how many organisations have built automation estates over time and now want AI agents to work alongside existing tools rather than replace them.

For operations leaders, a central issue is exception handling. Agent-based systems can introduce variability in outcomes, particularly when inputs are incomplete or when decisions must account for policy, risk appetite or regulatory requirements. Nintex says the phase-based approach treats exceptions as a natural part of execution and manages them within the broader process lifecycle.

Governance focus

Nintex Chief Product and Technology Officer Niranjan Vijayaragavan framed the release as an attempt to bring autonomy and control into the same design environment. "AI for modern business cannot be purely agentic, nor purely deterministic; it must support both," he said.

He linked that view to oversight and compliance in operational settings. "Our agentic business orchestration vision is built on enabling deterministic workflows and adaptive agents to operate within the same process framework, alongside human oversight. With Agent Designer and Orchestration, organisations can apply AI where judgment adds value while preserving control where precision and compliance matter most," he said.

Nintex also pointed to its AI UNLESS report, which found that 64 per cent of business leaders are embedding or consolidating AI into broader automation strategies to create unified platforms that orchestrate people, systems and AI agents. The company argued orchestration becomes more important as AI moves from isolated experiments into core operational processes, where governance requirements and system integration are typically stricter.

Partner reaction

Nintex partner AiGS - Ai Global Solutions described the release as a shift in how workflow automation products incorporate AI.

"This is a game-changer. Nintex Agent Designer represents the future of workflow automation, combining intelligence with governance in a powerful new way," said Kevin Schall, CEO, AiGS - Ai Global Solutions. "With capabilities such as supervisor agents, Nintex is enabling organisations to introduce advanced AI-driven automation while maintaining the oversight and trust that users expect from agentic AI systems."

Agent Designer and Orchestration are available in beta to select Nintex CE customers, with broader availability expected as testing progresses.