Incident Resolution with Digitate’s ignio AI Agent

At AI Field Day 7, Rahul Kelkar, Chief Product Officer at Digitate, presented the capabilities of ignio, an AI-based incident resolution agent designed to automate, augment, and improve IT operations. Ignio uses a logical reasoning model for incident resolution, leveraging enterprise blueprints to understand situations in a closed loop and applying automation where possible. When a fully automated response is not viable, ignio augments human efforts through assisted resolution, supplying prioritized incident lists based on business impact, providing situational context, and capturing both historical and episodic memory about recurring issues. The product integrates with various data sources to build a formal enterprise IT model, supporting information ingestion via templates or extraction from existing documentation, and includes adapters for common ITSM systems like ServiceNow for seamless change management.

Technically, ignio’s core incident resolution operates via automated root cause analysis, performing real-time health checks across application hierarchies—such as applications running on Oracle databases hosted on Red Hat servers—and comparing the current state to baselines to isolate anomalies. It can autonomously apply prescriptive fixes, such as restarting services, and then validate remediation by rechecking stack health. In more complex scenarios, like SAP HANA environments or intricate batch job dependencies in retail order management, ignio handles non-vertical, multi-layered issues involving middleware, business processes, and interdependent bad jobs. The solution features out-of-the-box knowledge for common technologies and allows continuous augmentation with customer-specific logic. Custom operational models and atomic actions can be enhanced using Ignio Studio, and the system learns from user feedback through reinforcement learning, improving accuracy in prioritizing incidents, suggesting fixes, and predicting service level agreement (SLA) violations before they occur.

Ignio extends beyond deterministic resolution to assist engineers and SREs via conversational augmentation. For issues not resolved autonomously, ignio provides contextual insights—including previous incidents, typical resolutions, and guidance for next-steps—while collaborating via a “resolution assistant” so humans can contribute domain knowledge and validate procedure. The demo showed proactive recommendation capabilities, identifying dominant recurring SLA violations and offering actionable, prioritized problem management insights. Ignio integrates with multiple agent-based platforms for orchestrated, multi-channel incident management flows, including email, Slack, and ticketing systems, using orchestration protocols and adapters. The platform employs advanced anomaly mining and sequence analysis, allowing users to identify root causes not only within vertical stacks but also through complex temporal and conditional relationships across business functions, ultimately supporting predictive, reactive, and continuous improvement use cases in large-scale enterprise IT environments.