Two generative AI products for the enterprise: endpoint remediation and natural language to SQL.
An LLM-generated remediation product for enterprise endpoints, built by a team of 35, and a natural-language-to-SQL assistant that removed engineering from the campaign loop.
Generative AI for endpoint management, positioned against Microsoft Intune and Nexthink. Chat-based iteration with the user to capture a problem, then LLM-generated PowerShell remediation scripts with automated documentation, security and privacy assessment, naming, tagging and categorization. Three core generative AI features including predictive maintenance. Team of 35 engineers.
Projected outcomes at the time: 30% reduction in IT helpdesk tickets, 25% gain in asset management efficiency, 15% development acceleration. Every figure here was a projection, not a measured result.
A business user asks in plain English; the system returns English pseudocode and SQL that the user can edit; it executes the query and returns the data with a high-level multivariate analysis. Campaign cycle time fell from two months to one day and the dependency on engineering for every new campaign disappeared.
Migration from legacy identity management systems to modern asset services for device enrollment and management, reducing cost and improving scalability and accuracy.
| 2 mo → 1 day | Campaign cycle time, Campaign Management AI. |
| 30% (projected) | Reduction in IT helpdesk tickets, Device Scripts. |
| 25% (projected) | Gain in asset management efficiency, Device Scripts. |
Generative AI products ship when the generated output carries its own documentation and security assessment, the same principle that governs AI in banking.