OutSystems Agent Workbench Accelerates Enterprise Agentic AI Adoption

Pioneers like Thermo Fisher, Axos Bank, and Ascot Insurance have more than 5,500 AI agents in development, delivering measurable efficiency gains and optimizing workflows

OutSystems, the leading AI-powered low-code development platform, today announced strong early momentum for Agent Workbench, which builds and orchestrates intelligent AI agents across data sets, workflows, or systems. Following its general availability launch on September 30, enterprises across industries are already using Agent Workbench to accelerate agentic transformation, moving from pilots to production and orchestrating multi-agent workflows with built-in human oversight that delivers measurable business outcomes.

As showcased during today’s OutSystems ONE World Tour Miami conference, Agent Workbench is experiencing impressive growth, with more than 5,500 agents in development. A fast-growing community of nearly 1,500 certified developers is already building enterprise-grade agentic applications with Agent Workbench.

"The enterprise AI race isn’t won by faster experiments; it’s won by building trusted AI-powered systems that scale,” said Woodson Martin, CEO of OutSystems. “With Agent Workbench, enterprises are building an agentic future that connects data, managing the full agent lifecycle, and orchestrating human-AI collaboration to transform how work gets done and deliver unprecedented business value. The early momentum we’re seeing proves that enterprises are ready to move beyond experimentation and deploy AI that truly performs at scale."

From Pilots to Production: Enterprises Are Going Live with Agentic AI

• Axos Bank is using Agent Workbench to accelerate operational efficiency and automate critical internal workflows. The team developed an intelligent log analysis agent that interprets error logs and provides real-time recommendations, reducing the need for manual analysis. They also built an automated document mapping agent to improve accuracy and eliminate repetitive data entry tasks.
• Thermo Fisher Scientific is transforming its customer operations with a Customer Escalation Agent built in Agent Workbench. The solution interprets unstructured data from customer interactions to automate triage and accelerate resolution, reducing manual intervention and improving time to resolution.
• Ascot Insurance is leveraging Agent Workbench to instantly assess critical Key Performance Indicators (KPIs) within their “Avail Claims” system. Where traditional systems fail to read unstructured data, the new workflow utilizes AI to intelligently interpret agent activity logs and notes to accurately determine KPI status—data previously inaccessible without Agent Workbench.
• MAGnet Auctions developed an AI agent within OutSystems to dramatically streamline the vehicle entry quality assurance process. The agent automatically identifies and reads odometer photos, comparing the detected mileage against the database record to flag potential discrepancies. By introducing this intelligent automation, MAGnet reduced the number of photos requiring manual human review by over 90%, delivering significant cost savings, improved data accuracy, and an enhanced customer experience. The initiative also enabled the reallocation of human resources from repetitive QA tasks to higher-value work — a clear example of how agentic AI can transform operational efficiency.

"Our work with OutSystems on Agent Workbench is a significant step in our AI strategy. We’re expanding the use of AI capabilities within OutSystems to realize immediate gains, without needing to invest in specialized AI roles,” said Kevin Hearn, SVP, Head of Consumer Bank Development at Axos Bank. “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. 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."

An Open Ecosystem: Deploying Agents Where Work Happens

OutSystems is building an open agentic ecosystem that allows enterprises to deploy AI agents across critical workflows and lets enterprises plug in any LLM and put it to work where it matters.

Connect a model once in Agent Workbench and reuse that connection across multiple agents and applications without rewrites or reconfiguration. With support for Bedrock, Azure OpenAI, Anthropic, Gemini, Cohere, Mistral, Databricks, AI2, IBM watsonx, Vertex AI custom models, and Hugging Face open source options, teams can route, mix, or swap models per use case to balance accuracy, cost, and latency while staying flexible as the AI landscape evolves.

On top of this open LLM layer, teams can tap into a fast-growing library of ready-to-use agents and agentic applications for the systems where work happens every day:

• Productivity and scheduling platforms such as Google Calendar, where smart schedulers create meetings, set agendas, and capture follow-ups.
• Collaboration and knowledge systems like Confluence, and Notion, where agents post updates, summarize channels, and send reminders to accelerate decision-making.
• Work management tools including Trello, Asana, Monday.com, and Jira, where agents automate intake, create or update tasks, assign owners, and track progress.

As part of its mission to make enterprise AI adoption accessible, OutSystems has trained and certified nearly 1,500 AI developers on Agent Workbench, enabling organizations to rapidly scale their agentic capabilities. As OutSystems expands its global enablement programs, enterprises are encouraged to onboard their AI developers before the end of 2025.

Agent Workbench is now generally available worldwide. To learn more about how enterprises are using agentic AI to drive innovation, visit the OutSystems Agent Workbench homepage.