OutSystems, a leading AI-powered low-code development platform, has announced the general availability of its Agent Workbench, a new solution designed to enable enterprises to create and orchestrate intelligent AI agents across various use cases, departments, data sets, and workflows.
The Agent Workbench aims to address the significant market need for AI agent deployment, as 93% of organizations prioritize their development but often face challenges with governance, security, integration, and scalability. The platform is designed to unify data, workflows, experiences, and intelligent agents into a secure, controlled, and trusted enterprise-grade system.
Hundreds of customers and partners, including Axos Bank, Thermo Fisher Scientific, The Arch Company, Grihum Housing Finance, and TeamWork, participated in the Agent Workbench Early Access Program. These participants reported notable improvements in operational efficiency, accelerated speed to value with agentic AI, and enhanced digital experiences for both customers and employees.
Kevin Hearn, SVP, Head of Consumer Bank Development at Axos Bank, commented on their experience: “We plan to expand the use of AI capabilities within OutSystems to realize immediate gains, without needing to invest in specialized AI roles. With Agent Workbench, we can quickly and safely create agents for specific use cases, such as analyzing error logs and automating data entry from documents. Our work with OutSystems on Agent Workbench is a significant step in our AI strategy. By creating and embedding agents into our operations, we are building a more intelligent and responsive banking ecosystem that is ready for the future of finance.”
The announcement was made live at the OutSystems 2025 ONE Conference in Lisbon, Portugal, an annual event that gathers IT leaders, developers, and partners. OutSystems CEO Woodson Martin highlighted agentic AI as a powerful tool for enterprise innovation, emphasizing its potential to drive revenue, enhance operational efficiency, and deliver superior customer experiences.
During the Early Access Program, participants demonstrated a range of impactful applications:
Axos Bank, a technology-driven financial services company, optimized internal operations by automating complex manual tasks. This included developing an intelligent log analysis agent to interpret error logs and provide recommendations, as well as automating data entry from documents, leading to improved data accuracy and employee efficiency.
Thermo Fisher Scientific transformed client operations through a Customer Escalation Agent designed to interpret unstructured data from customer interactions. This eliminated the need for manual triaging, accelerating problem resolution and reducing the workload on support teams.
In the United Kingdom, The Arch Company, a leading small business landlord, utilized Agent Workbench to implement a content classification agent that intelligently routes customer service inquiries, ensuring faster and more accurate responses. This supports their comprehensive customer and asset management system.
TeamWork, an international consulting and technology integration group, developed a tool for its support teams. This tool provides quick access to ticket information and real-time guidance while using AI agents to automate the resolution of routine tickets, allowing teams to focus on higher-impact issues and anticipate a considerable reduction in resolution times.
Grihum Housing Finance, an affordable housing finance company in India, is deploying Agent Workbench to enhance the accuracy of loan underwriting and streamline property evaluations. Specialized agents will analyze property evaluation reports and suggest technical property deviations during the loan origination process, transforming a previously manual and error-prone workflow.
The generally available version of Agent Workbench includes enhanced features such as an agent marketplace and Model Context Protocol (MCP) support, which enables agents to directly access enterprise systems, external tools, and services. The platform now supports a broad range of Large Language Models (LLMs), including AWS Bedrock, Azure OpenAI, Anthropic, Gemini, Cohere, Mistral, Databricks, AI2, and IBM’s WatsonX, alongside custom-built models on VertexAI and open-source models on HuggingFace.
Hélio Pimenta, Associate Partner, KPMG Low-Code Center of Excellence, noted, “Agent Workbench delivers the speed-to-value and governance guardrails that organizations have come to expect from OutSystems development. It enables teams to design, orchestrate, and deploy AI agents that operate autonomously – from planning tasks to integrating with enterprise systems – while maintaining control through policies, auditability, and compliance across the lifecycle.”
OutSystems, founded in 2001, provides an AI-powered low-code development platform used by thousands of global customers. The company’s network serves over 60 million end users, with more than 500 partners and active customers across 75+ countries and 20+ industries.
