Agentic AI for IT Operations Case Study

Problem Statement

One of the Leading global credit rating firm had their L1 Operations relied heavily on a 24×7 team to manually monitor scheduled jobs and applications, perform application sanity checks, execute recurring and ad-hoc operational activities, and respond to user incidents. As the technology landscape and number of supported applications grew, the operating model required additional human capacity to absorb additional workload.

This created a linear scaling challenge, where growth in applications and operational activity translated directly into increased staffing requirement. The objective was to redesign L1 operations so that routine operational activities could be executed automatically, with human intervention focused primarily on exceptions and activities requiring judgment.

Challenges

Manual & Human Dependency

Continuous monitoring required round-the-clock staffing. A significant portion of monitoring, validation, recovery, and operational procedures required analysts to execute repetitive SOP-driven steps. The L1 function was not limited to monitoring. It included scheduled job monitoring, application sanity testing, recurring and ad-hoc activities, and handling selected user incidents.

Linear scaling of operations

Adding applications, jobs, integrations, and operational responsibilities increased the workload on the L1 team and created pressure to add additional capacity.

Solution

Stratacent analyzed the L1 operational workload and redesigned the operating model around automation-first execution.

Rather than applying a single automation technology, activities were assessed based on their characteristics and automated using the most appropriate approach:

Deterministic operational procedures, integrations, monitoring activities, validations, and repeatable recovery actions were automated through custom-developed solutions. Certain workflow-driven and rules-based operational activities were automated using RPA, including activities previously dependent on manual execution.Activities requiring contextual interpretation, decision logic, or multi-step orchestration were addressed through agentic solutions capable of analyzing operational information and determining the appropriate next action.

Automation was implemented across the major areas of L1 responsibility:

Job Monitoring → Application Sanity Testing → Recurring & Ad-Hoc Operations → User Incident Handling 

The resulting model shifted L1 operations from humans executing routine operational procedures to automation executing the procedures and humans managing exceptions where necessary.

Value Delivered

Enabled elimination of the dedicated L1 operating layer

The reduction in manual workload allowed the credit rating firm to transition away from a dedicated L1 team and operate with a more automation-led support model. Reducing dependency on L1 staffing lowered the ongoing cost of delivering operational support.

Removed linear scaling constraints

Growth in applications and operational workloads no longer requires a proportional increase in L1 staffing, creating a more scalable operating model.

Improved reliability and consistency

Automated execution reduces dependency on individual analysts and ensures repeatable operational procedures are performed consistently. Automation removes many of the missed steps and execution errors inherent in repetitive manual processes.

Established the foundation for autonomous operations

The combination of traditional automation, workflow automation, and agentic capabilities creates a path toward increasingly autonomous and eventually self-healing operations.

Transforming L1 Operations through Automation and Agentic AI - reducing manual effort, improving reliability, and enabling scalable operations.

Manual & Human Dependency

Linear scaling of operations

Your Challenge. Our Expertise.