Certified AI colleagues have completed preliminary construction, possessing verified integration capabilities, professional skills, and enterprise-level control, and are capable of undertaking operational roles in the fields of IT and security.
San Francisco, November 6, 2026 / PRNewswire / -- Today, we announce the launch of our AI Collaborator Market ( AI Coworker Marketplace ), which enables enterprises and security teams to utilize pre-built, certified AI collaborators designed to undertake operational roles in areas such as cloud infrastructure, access management, security, and device operations.
The market has introduced AI colleagues certified by Atomicwork, and provides verified integration and skill coverage for the systems required by their roles. Each AI colleague has also been assessed based on real-world corporate usage scenarios, enabling organizations to deploy AI workers more quickly, without having to build and verify from scratch.
The first batch of products includes employee support operations by colleague AI, as well as Cloud Ops Engineer, Access Ops Manager, Network Ops Engineer, Sec Ops Manager, and Device Ops Engineer. These colleagues can work within the enterprise systems that also include AWS, Microsoft Azure, Google Cloud Platform, Okta, Active Directory, Splunk, and Datadog.
Atomicwork, the Chief Executive Officer and Co-Founder, Vijay Rayapati stated: "The IT team of the enterprise does not need another proxy directory that can only perform a limited number of demonstrations. What they need are AI colleagues who can be trusted and are capable of undertaking real operational tasks within the systems that the enterprise's business relies on. We have built these AI colleagues around specific IT and security roles, equipped them with the tools and skills required for these roles, and assessed them based on real corporate scenarios. As a result, customers can have more confidence when deploying them, using them as their eyes, ears, and hands when additional support is needed."
Build AI labor force for enterprise IT
Atomicwork enables service teams to design, deploy, and manage AI colleagues who undertake operational roles rather than single tasks. Each AI colleague can have their own operating procedures, tools, permissions, reporting structures, and budgets. Organizations can also use many of the same operational metrics as human teams to measure their work, including average resolution time and the number of tickets processed.
Building on this model, the market has further expanded to provide customers with Atomicwork colleagues who have already been built for specific operational fields. Certification means that each AI colleague possesses the integration and skills required for their designated role and has been evaluated based on real-world business use cases.
For example, Cloud Ops Engineer can operate between AWS, Azure, and Google Cloud Platform, performing diagnostics, event routing, and identifying gaps in the cloud environment. It combines the diagnostic capabilities of a specific platform with upgrade handling and problem-solving skills, and leverages the organization's existing skill sets and code libraries.
Access Ops Manager allows for cross-enterprise identity and system access, helping team administrators process employee access requests and automate repetitive permission granting processes. Sec Ops Manager supports secure operations on the enterprise security platform, while Device Ops Manager focuses on device-related IT operations.
Each AI colleague is equipped with the skills and integrations required for their role, thereby reducing the amount of work that the IT team would otherwise have to spend on connecting systems, defining processes, and verifying proxy capabilities.
Provide enterprises with a faster path to establishing credible AI operations.
Cloud infrastructure, security operations, and network operations require specialized knowledge and generate a large amount of repetitive work. When enterprises introduce AI into these environments, they must also determine which systems the agents can access, what operations they can perform, and how these operations should be governed.
Most proxy markets only provide a list of available proxies, but there is limited evidence regarding what these proxies are actually capable of doing. Atomicwork indicates that its certification process fills this gap by verifying whether each AI colleague possesses the necessary integration skills and expertise for their operational role, and ensuring that they have undergone review in real-world business scenarios.
Atomicwork also provides governance controls over network access, tools, and code execution to help organizations manage the way AI colleagues interact with enterprise systems.
Customers can use different AI models and proxy runtime frameworks to deploy certified AI colleagues, or they can also deploy them within their own infrastructure. After deployment, organizations can evaluate the performance of these AI colleagues when handling tickets and view weekly recommendations for improving their skills and operational processes. These recommendations will compare actual performance with expected behavior and identify areas where AI colleagues may need improvement.
Atomicwork also supports the Cross App Access (XAA) protocol of Okta. This is an extension of OAuth, aimed at achieving secure proxy-to-application and application-to-application connections. XAA enables AI colleagues to perform operations on behalf of authorized users, while maintaining corporate access control.
Early deployment has already been carried out in the IT enterprise scenario.
Companies are already deploying Atomicwork AI colleagues to automate repetitive IT operations. The company states that after Atom, Access Ops Manager is the second-largest certified AI colleague in terms of deployment volume, capable of assisting organizations in handling employee access requests and permission activation processes. Among customers using the automated Atomicwork for access management, the median turnaround time for applications that do not require approval is only 18 seconds, while for those that do require approval, it is 2.5 hours.
This market represents the first phase of Atomicwork's broader plan, which is to create an ecosystem for the colleagues of enterprise AI. The initial market is composed of colleagues from Atomicwork who have built and certified it, while third parties can contribute their skills and integrations through their MCP stores and skill registries.
Over time, Atomicwork plans to expand this market, allowing third-party organizations to publish their own AI colleagues, and to verify them based on the Atomicwork corporate evaluation criteria.
About Atomicwork
Atomicwork is a AI native ITSM and ESM platform, which also provides the AI workforce necessary to operate this platform. It is specifically designed for companies that aim to scale through intelligence rather than merely relying on expansion of personnel. Its governed AI colleagues can manage end-to-end access control, incident response, employee onboarding, and other enterprise service delivery projects. From AI grassroots growth-oriented companies to global giants, various organizations are using Atomicwork. The company is headquartered in Palo Alto, California, and its investors include Khosla Ventures, Battery Ventures, and Peak XV, among others.
For more information, please visit atomicwork.com.
