Enterprise · Service Operations
Hands-free ticket triage with cooperating AI agents.
Fortune 10 healthcare enterprise
A multi-agent system that validates incoming ServiceNow tickets against master client records, applies turnaround-time rules, and assigns each ticket to the right person — automatically, end to end.
Enterprise · Service Operations
Challenge
What was the challenge?
Every incoming service ticket had to be processed against three reference sources before work could begin: the ticket's details and attachments validated against a master intake list holding client-specific rules, a turnaround-time grid consulted to set the right commitment, and an assignment grid checked to find the right owner.
It was rules-driven work performed entirely by hand — slow, repetitive, and easy to get wrong, with every mistake surfacing later as a missed SLA or a ticket sitting in the wrong queue.
Approach
How did we approach it?
The manual process was already a pipeline of discrete decisions, so we built it as one: four specialized agents, each owning exactly one decision, cooperating on each ticket in sequence.
A Triage Agent first identifies tickets eligible for immediate assignment without validation, so simple cases skip the queue entirely. The remaining tickets flow through an Intake Master Agent that matches the client to the master intake list and pulls that client's specific rules, a TAT Agent that references the turnaround-time grid, and an Assignment Agent that selects the owner from the assignment grid and writes everything back to ServiceNow — assignee, TAT, and validation results — closing the loop with no human touch for standard cases.
Delivered
What did we deliver?
- A Triage Agent that identifies tickets eligible for immediate assignment without validation.
- An Intake Master Agent that matches each ticket's client against the master intake list and applies that client's specific rules to the ticket and its attachments.
- A TAT Agent that sets the correct turnaround-time commitment per ticket from the TAT grid.
- An Assignment Agent that selects the right assignee from the assignment grid and updates the ServiceNow ticket in place with assignee, TAT, and validation results.
Mechanism
How does it work under the hood?
- 01Incoming tickets and their attachments are parsed and validated against the intake master containing rules for every client.
- 02The four agents run as a pipeline — triage, intake matching, TAT lookup, assignment — each responsible for exactly one decision, which keeps every step explainable.
- 03Fast-path tickets identified at triage skip validation and go straight to assignment.
- 04The Assignment Agent performs the final write-back to ServiceNow, so the ticket record itself carries the full result of the automated process.
Outcome
What was the outcome?
The proof of concept demonstrated end-to-end, hands-free processing for standard tickets: validated against client rules, given the right TAT, and assigned to the right person with the ticket updated automatically.
Because each agent owns a single decision, the system stays auditable — for any ticket, you can see exactly which rule fired at which step, a property that matters when automation touches client SLAs.
Learnings
Key learnings
Rules-driven manual work is the ideal first target for agents — the decisions are already explicit, they just need to be executed consistently.
One agent per decision keeps the system auditable: for any ticket you can see exactly which rule fired at which step, which matters when SLAs are involved.
A fast-path triage step protects throughput — simple cases skip validation entirely instead of queueing behind complex ones.
Related stories
Healthcare · Document Intelligence
Agentic AI that reads tables no algorithm could
A multi-agent LLM system — an extraction agent, a standardization agent, and a self-checking feedback loop — that digitizes contract tables too complex for traditional parsing: borderless layouts, merged cells, tables spanning pages.
Enterprise · Proposal Operations
Answering RFP questionnaires with agentic RAG
An agentic retrieval-augmented system that answers standard RFP questions instantly from a library of expert-approved responses — and intelligently routes novel questions to the right team.
Work with us