Can OutSystems' Agent Workbench Bring Agentic to the Masses?
OutSystems has formally unveiled Agent Workbench, a fresh platform created to make agentic AI accessible to everyone.
Over recent years, the firm has gained growing recognition for its low-code AI solutions that enable organisations to build and tailor AI models without requiring specialist engineering support.
Agent Workbench, which was revealed for general availability at OutSystems' yearly conference in Lisbon this week, extends this same approach to agentic AI, one of the standout technologies of 2025.
Through AI agents, organisations can essentially expand their workforce via the power of automated digital work.
"Organisations are excited by the promise of AI and agentic systems, but are struggling with endless pilots and ungoverned tool sprawl while the business impact stalls," explains Woodson Martin, CEO of OutSystems.
"Legacy systems, siloed data, fragmented AI tools and complex AI development cycles are slowing progress.
"We built Agent Workbench to make it possible to unlock custom agents as a true business enabler, not an experiment," he says.
Major companies trial Agent Workbench
Axos Bank took part in the early access initiative for Agent Workbench and continues to utilise the platform for automating log analysis and document processing operations.
"We plan to expand the use of AI capabilities within OutSystems to realise immediate gains, without needing to invest in specialised AI roles," says Kevin Hearn, SVP and Head of Consumer Bank Development at Axos Bank.
Axos is additionally developing agents for error log analysis and automating document data entry.
Thermo Fisher Scientific represented another major corporation to trial Agent Workbench during its initial phase.
Using OutSystems' new offering, the company implemented a customer escalation agent to process unstructured data from customer interactions and remove manual triaging procedures.
Meanwhile, Arch Company, Britain's biggest small business landlord, introduced a content classification agent that has enabled it to direct customer service queries without manual involvement.
These pilot programmes enabled OutSystems to demonstrate its technology's capabilities to significant clients, whilst also providing the team with early feedback from enterprises with substantial operations which proved to be invaluable testing.
Tackling governance and integration obstacles
For Woodson, the introduction of Agent Workbench responds to the common difficulties that most enterprises have encountered with agentic AI deployment thus far.
The firm referenced research showing that 93% of organisations are making AI agent development a priority but encounter obstacles regarding governance, security, integration and scalability.
"Agentic AI is one of the most powerful levers for enterprise innovation and transformation, empowering organisations to grow revenue, drive operational efficiency and deliver amazing customer experiences," he explains.
Through this new platform, Woodson and colleagues aim for businesses to streamline the entire process, consolidating their data, workflows and intelligent agents into a single trusted platform.
Technical features and model compatibility
The generally available release of the Workbench features an agent marketplace and Model Context Protocol support, enabling agents to connect with enterprise systems and external tools directly.
Agent Workbench is compatible with models from AWS Bedrock, Azure OpenAI and Anthropic, along with newer LLMs including Gemini, Cohere, Mistral, Databricks, AI2 and IBM's watsonx.
The platform additionally supports custom-built models on VertexAI and open-source models on HuggingFace, ensuring integration with existing systems is straightforward.
Enterprise applications take shape
TeamWork, an international consulting firm, created a multi-agent system that delivers real-time guidance to support teams whilst automating routine ticket resolution.
The firm expects that a significant proportion of tickets will be resolved automatically, with resolution times for complex cases projected to drop substantially.
Grihum Housing Finance in India is implementing the platform to enhance loan underwriting accuracy by examining property evaluation reports and identifying technical property deviations.
KPMG's Low-Code Centre of Excellence participated in testing the platform during its early access period.
"Agent Workbench delivers the speed-to-value and governance guardrails that organisations have come to expect from OutSystems development," says Hélio Pimenta, Associate Partner at KPMG's Low-Code Centre of Excellence.
