Banking & Financial ServicesRegional bank, ~3,000 employees

Automating claims triage for a regional bank

An anonymized engagement where a regional bank replaced a fragmented claims-triage process with a private, governed AI workflow on Vestval Flow.

Tools / product used · Vestval Flow + Custom AI agents

Challenge

Claims triage was handled across four legacy tools, three spreadsheets and two outsourced teams. Average time-to-first-touch was measured…

Solution

A private deployment of Vestval Flow with custom AI agents for intake classification, document extraction and routing — backed by a clean…

Architecture

We ran a four-week paid discovery, mapped the actual claims path (not the documented one), instrumented the existing flow to get baseline…

Timeline

4-phase implementation · Vestval Flow + Custom AI agents

Impact

Time-to-first-touch reduced from days to hours (qualitative; specifics confidential)

Challenge

Claims triage was handled across four legacy tools, three spreadsheets and two outsourced teams. Average time-to-first-touch was measured in days, not hours. Audit posture was weak and analyst leverage was poor.

Objectives

Reduce time-to-first-touch on inbound claims from days to hours
Establish a defensible, regulator-ready audit trail across every triage decision
Lift analyst leverage without changing headcount
Retire two outsourced workflows without service disruption

Approach

We ran a four-week paid discovery, mapped the actual claims path (not the documented one), instrumented the existing flow to get baseline numbers, and proposed a governed agent architecture with humans-in-the-loop at every legally meaningful step.

Solution

A private deployment of Vestval Flow with custom AI agents for intake classification, document extraction and routing — backed by a clean audit log, role-based access, and human approval gates. Integrations with the bank's existing core and CRM.

Implementation approach

1
Discovery & baseline instrumentation
Four weeks of paid discovery — process shadowing, instrumentation of the existing flow, and a measured baseline that every later metric is compared against.
2
Governance design before build
Maker-checker model, audit log schema, role matrix and escalation rules signed off by compliance and legal before a single agent was written.
3
Phased agent rollout
Intake classification first, document extraction second, routing third — each behind a human gate before being promoted to default.
4
Parallel running and cutover
Eight weeks of parallel running with legacy queues kept live, decision-by-decision comparison, then phased cutover by claim type.

Technologies used

Vestval Flow
Private LLM deployment
RAG over policy documents
Document extraction pipeline
OIDC + role-based access
Immutable audit log

Outcomes

Time-to-first-touch reduced from days to hours (qualitative; specifics confidential)
Analyst leverage materially improved — same team handling significantly more volume
Cleaner audit trail across every triage decision
Compliance and legal sign-off retained at every decision point

Lessons learned

Baseline instrumentation before any build is the single most valuable phase — without it, ROI is debatable forever.
Maker-checker encoded in workflow definitions, not in convention, is what makes BFSI AI defensible.
Parallel-run windows feel expensive and are absolutely worth it.