The Pulse

NFON makes AI Agent Studio available for business automation

NFON says AI Agent Studio is available for companies to create and test custom AI agents. The workspace combines natural-language development with visual workflows, while some system integrations require technical implementation.

AI.info Team · · fact-checked · against 4 sources ↓ · Published 12 Nov 2026

Businesses can now use NFON’s AI Agent Studio to build and test custom AI agents through a conversational interface or visual workflows. The Munich-based communications company said in a November 12 release that the workspace is available for automating business tasks.

The product adds a conversational approach alongside the existing visual method for creating AI applications. NFON had already described the studio and its Agent Mode in an October 2 blog post.

Building an agent

Agent Mode lets users describe an agent’s task, behavior and requirements in natural language. The system then generates an initial working configuration that users can test and refine through dialogue.

Proposed changes are reviewed and approved before they are applied, and changes can be traced and reversed. The AI Assistant can be asked to alter instructions, add requirements or adjust an agent’s behavior.

In Flow-Based Mode, users create visual workflows and configure individual process steps. They specify which actions should take place under particular conditions, providing rule-based control over the process.

The workspace uses botario technology and brings creation, testing and refinement together. According to NFON, separate development and production endpoints support controlled testing while an agent is being refined.

From demo to live systems

The company says simple agents with little or no third-party integration can follow a self-service path. Complex APIs, third-party systems and individual production designs require technical implementation, supported by qualified partners and NFON Professional Services.

A demo can use simulated connections; real system integration is handled separately.

NFON gives insurance claims as one example: an agent identifies the type of claim, collects details and forwards sensitive cases to a designated team. The company also describes examples for answering public-service questions and coordinating appointments, and for handling address changes and meter readings in energy customer service.

In each example, connecting the agent to the relevant live customer-management, appointment, billing or metering system is a separate step. The demonstrations show the intended service tasks rather than a completed connection to those systems.

Partner distribution

NFON is making the studio accessible through its European partner network. Sales partners can use it to develop customized automation for customers in different sectors, test ideas, demonstrate them in customer meetings and refine them with clients.

NFON says existing voicebots already receive insurance claims, process meter readings for energy suppliers and pre-qualify customer inquiries in tradespeople’s workshops. Its partners can build on those uses to automate additional process steps.

The company is supporting the introduction with training and certification programs, implementation examples and information materials for customer consultations. An interactive product demonstration is also available.