Outsystems Unveils Open Agentic Systems Platform
OutSystems today introduced the Agentic Systems Platform and the OutSystems Agent Experience, a new platform layer that exposes a suite of A2A and MCP tools and services..
“Enterprise leaders are realizing they need leverage and control in this highly fragmented and dynamic AI market. To maintain that leverage and protect margins, organizations must separate their proprietary business logic and data from specific AI providers,” said Woodson Martin, CEO at OutSystems. “Our Agentic Systems Platform provides the open and neutral platform to ensure optionality and control. Customers can optimize for digital sovereignty, performance and cost and can work across all the latest models and tools without rebuilding their core operations every time. In a time of great change, we are ensuring the CEO and CFO maintain control over the company’s future.”
Built on the Agentic Systems Platform, which is powered by the OutSystems Enterprise Context Graph, the following innovations fall into three domains: Agentic Systems Engineering, Agentic Enterprise Orchestration, and Agentic Industry Solutions.
Agentic Systems Engineering: OutSystems today expanded its Agentic Systems Engineering capabilities through new Agent Experience development services – a native integration with Kiro (an Agentic Development Environment from AWS) to build, manage, and govern agents and applications on the OutSystems platform. OutSystems also launched Legacy Modernization Services powered by AWS Transform, delivering an automated pipeline that migrates legacy systems based on technologies like COBOL and Lotus Notes into high-performance agentic systems. Kiro development services, as well as Legacy Modernization Services, are available in preview today for select customers with expanded AWS Transform integration capabilities coming in Q3.
Agentic Enterprise Orchestration: The company today unveiled Agentic Enterprise Orchestration, powered by the Enterprise Context Graph and featuring Amazon Bedrock. This is the next generation of Agent Workbench, enabling organizations to architect, orchestrate, and govern an agile workforce of AI agents. Agentic Enterprise Orchestration launched today, bringing advanced agent evaluations, precise guardrails, integrated semantic search, and deepened Amazon Bedrock support to the enterprise. This combines the OutSystems dynamic operational context layer with Amazon’s broad model management tooling to provide access to the most cost-effective and high-performing models. Available today, existing customers can leverage these new GA capabilities using Agent Workbench.
Agentic Industry Solutions: OutSystems today also launched its first end-to-end Banking Solution for Loan Origination, the first of a family of upcoming Agentic Industry Solutions. By streamlining the deployment of cutting-edge agentic technology, these Solutions de-risk AI adoption while radically accelerating time-to-value. Also leveraging Amazon Bedrock, Agentic Industry Solutions benefit from access to a library of domain-specific models (e.g. trained on finance, law, or use case specific problems like document processing) to further optimize performance and costs at scale. Interested developers can get started today with the OutSystems Banking and Financial Services Agent Kit, with the first end-to-end Banking Solution available early Q3.
Testlio Launches Human-in-the-Loop Testing for AI Agents
Crowdsourced testing platform provider Testlio today has launched AI Agent Testing services , which brings together the company’s human-in-the-loop orchestration and proprietary AI-driven insights from 13+ years of QA expertise into agentic workflows.
An agent that performs in a controlled run can still break the moment it meets a real device, a regional payment flow, or a user who doesn’t follow the script. Testlio’s AI-certified global testing community validates agentic workflows in the unpredictable conditions where failures actually occur. Testlio tackles the most critical challenges teams face today with their agentic workflows:
-
Agents fail across unpredictable real-world cases. Testlio’s AI-certified testers run thousands of what-if scenarios to keep agents contextual and brands out of trouble.
-
Automation-only evaluations miss nuanced issues. Testlio’s proprietary evaluation framework captures the subtle risks and behavioral issues that automation overlooks.
-
Agentic failures show up as quiet liabilities and policy violations. Testlio verifies data moves accurately between steps, catching critical leaks before customers do.
-
Companies that release fast are often at the highest risk of losing years of trust. Testlio matches rapid deployment cycles with high-velocity, real-time insights.
“Testlio is uniquely positioned for this moment, combining years of proven global payments testing with this new AI agent validation service,” said Summer Weisberg, CEO at Testlio. “As enterprise AI agents evolve from handling simple customer queries to executing complex agentic payment workflows, where AI is granted digital wallets to independently authorize transactions, organizations need a partner that goes beyond automated simulation. They need access to humans with real, localized payment methods to act as a safety net, alongside experts specializing in AI agent evaluation. That’s exactly what Testlio provides.”
Codex for every role, tool, and workflow
Codex is introducing plug-ins that enable non-developers to build apps, create dashboards and briefs, and more. While Codex was created as a tool for software development, non-developers now make up about 20% of overall Codex users and are growing more than 3x as fast as developers.
To meet these news roles where people work. Codex today announced plug-ins that can adapt Codex to whatever your roles and tools are, along with annotations for refining the AI results.The company is also previewing the ability to create shareable interactive websites and apps.
OpenAI said in its blog announcement that the six new role-specific plugins that make Codex useful for more kinds of knowledge work, requiring no coding, are:
- Each role-specific plugin(opens in a new window) bundles the relevant apps, skills, instructions, and workflows. Together, they include 62 popular apps and 110 skills.
- The data analytics plugin(opens in a new window) helps analysts and business teams answer questions with data. They can explore product and business data, explain why key metrics changed, and create reports and dashboards using tools like Snowflake, Databricks Genie, Hex, and Tableau, with more coming soon.
- The creative production plugin(opens in a new window) helps marketing and creative teams turn a brief into assets they can review. Teams can create campaign boards, make and refine display ad variations, and produce product lifestyle shots or ecommerce-ready image sets with tools like Figma, Canva, Shutterstock, Picsart, and Fal.
- The sales plugin(opens in a new window) helps sales teams bring customer context into the work that moves deals forward. Sales teams can find high-priority accounts and signals, prepare for customer meetings, complete follow-ups, update customer records, build close plans, and review deals at risk using tools like Salesforce, HubSpot, Slack, Outreach, Clay, Rox, and Actively.
- The product design plugin(opens in a new window) is built for turning early ideas into prototypes teams can review. Teams can explore product directions, audit user flows, prototype from a live URL, and make static screenshots interactive, with work that can be carried forward in tools like Figma and Canva.
- The public equity investing plugin(opens in a new window) helps investors make sense of market and company information. They can review earnings, compare companies, track signals, and assess whether an investment thesis is strengthening or weakening using information from Moody’s, Daloopa, Datasite, FactSet, LSEG, S&P, PitchBook, and Hebbia.
- The investment banking plugin(opens in a new window) helps bankers turn research and diligence into client-ready materials. They can prepare pitch materials, analyze comparable companies and transactions, and turn diligence into recommendations using trusted data.
Neo4j Acquires GraphAware to Launch Intelligence Analysis
Graph intelligence platform provider Neo4j announced it has an agreement to acquire intelligence analysis software company GraphAware, which Neo4j said will launch next-generation graph technology solutions powerd by AI and based on open standards.
The work, it said, will serve as “a proven and trusted alternative to Palantir Gotham. Bringing GraphAware’s intelligence analysis capabilities into Neo4j’s graph intelligence platform will give government agencies a powerful, sovereign alternative for collaborative investigation and explainable decision intelligence across complex, connected data. This will put customers in control of their deployment, data, and exit path, not the vendors who build the systems.
Advances in AI and growing geopolitical tensions have made sovereign solutions essential. Organizations need the ability to own, manage, and control their data in ways that best serve their needs, whether that means keeping it within national borders for compliance and security or freely enabling secure access to support emerging AI use cases and innovation.
The era of proprietary black-box solutions is over. Government agencies require the choice and control only made possible by an architecture based on a certified open standard that is modular by design and built to naturally integrate with all enterprise AI platforms.
Emil Eifrem, Founder and CEO of Neo4j, said, “For over a decade, governments have utilized intelligence analysis software from Palantir. Now they will have a legitimate choice with GraphAware Hume, powered by the Neo4j graph intelligence platform. We aren’t just launching a new intelligence analysis alternative to Palantir Gotham. We’re acquiring a proven solution that’s built on open-standards technology, and already trusted by government agencies throughout the world.”
