Expense management · Production

How Ramp Built LLM-Backed Expense Policy Agents Users Can Trust

The problem

Expense approval was a tedious, time-consuming task traditionally handled by managers, and applying LLMs in a finance product required careful design to avoid losing user trust through low-quality or unexpected outputs.

Workflow diagram · grounded in source
1
Expense submitted for approval
trigger
“approving expenses. This has traditionally been a manager's responsibility but we're now applying a "policy agent" that finance teams can trust to match or exceed human judgment”
2
Policy agent evaluates expense
ai_action
“the policy agent links directly to sections of the user's expense policy that its reasoning references”
3
Reasoning and citations output
output
“This first bullet point is output from an LLM explaining why the expense was approved. The info icon links directly to the section in the user's expense policy describing the requirements.”
4
Uncertainty escalation
routing
“When the agent isn't sure, we fall back to the pre-agent escalation process users are accustomed to”
5
Deterministic rule guardrails
validation
“We layer deterministic rules on top: dollar limits, vendor blocklists, category restrictions”
6
Feedback loop improves policy
feedback_loop
“This feedback loop not only reduces the amount of work humans need to do over time, but also improves policy accuracy”
Reported outcome

Since enabling the policy agent at Ramp, more than 65% of expense approvals are fully handled by the agent, with a feedback loop that reduces human workload and improves policy accuracy over time.

Reported metrics
Expense approvals handled by agentmore than 65%
Human workload over timereduces the amount of work humans need to do over time
Policy accuracyimproves policy accuracy
Reported stack
LLMworkflow builder
Source
https://builders.ramp.com/post/how-to-build-agents-users-can-trust
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Since enabling the policy agent at Ramp, more than 65% of expense approvals are fully handled by the agent, with a feedback loop that reduces human workload and improves policy accuracy over time.

What tools did this team use?

LLM, workflow builder.

What results were reported?

Expense approvals handled by agent: more than 65%; Human workload over time: reduces the amount of work humans need to do over time; Policy accuracy: improves policy accuracy (source-reported, not independently verified).

How is this expense management AI workflow structured?

Expense submitted for approval → Policy agent evaluates expense → Reasoning and citations output → Uncertainty escalation → Deterministic rule guardrails → Feedback loop improves policy.

Related expense management cases