In the initial phase, Gluware’s focus was on configuration management and drift detection. Their system was able to detect deviations from approved configurations in network devices and suggest solutions. However, manual approval was required for each remediation by network operations teams.
Moving on to the second phase, automatic remediation was introduced as customers grew more confident in the system. This allowed for automatic correction of devices that strayed from approved standards without the need for human approval. This phase marked the transition to self-operating networks where configuration drift triggered instant automated corrections.
The latest phase, represented by Titan, involves system-determined operations where AI identifies and implements necessary changes within defined risk parameters. Unlike drift remediation, which restores known-good states, Titan handles new modifications suggested by multiple AI systems that may conflict. The platform ensures coordination between various AI agents to prevent conflicts while maintaining network stability.
Titan’s architecture for network and agentic AI coordination comprises integrated components working together to address the multi-agent coordination challenge. The Intelligent MCP Server uses the Model Context Protocol to coordinate Gluware’s automation capabilities with external AI agents. The Gluware Agent executes automation tasks, while the Co-Pilot offers a user-friendly interface for network operations teams. The MCP Server includes a validation engine to verify actions before execution, allowing third-party agents to request network changes while maintaining control over execution. Initial integrations include NetBox and ServiceNow.
