NodeFox
Case processing

Hand your examiners cases that are ready to decide.

NodeFox classifies each case’s attachments, pulls the policy and claim record, and checks the file for missing or conflicting evidence. Complete files go to the decision queue. Gaps and contradictions reach a reviewer with the source pages side by side. Built on NodeFox’s executable workflow platform.

How it runs

One workflow, 3 different cases.

Pick a case to see which steps run, what a reviewer is shown and what gets written.

Synthetic example · Case file received

Case readiness review

Arrives

Case 60418 · Ridgefield Cabinetry · water damage

Loss date
Nov 2, 2026
Policy period
Jan 1 – Dec 31, 2026
Repair estimate
$18,450
Photos and notes
Attached

Runs

  1. Claim form and attachments from the intake inbox, a portal export or a watched folder.

  2. Claim form, estimates, invoices, reports and correspondence: dates and amounts into a typed record.

  3. Case system API: coverage period, limits and prior claims on the policy.

  4. Exact rules in code: required evidence present, dates inside coverage, amounts agree across documents.

  5. Every required document is present, and dates and amounts agree.

  6. The missing or conflicting item, the documents involved and their page references.

  7. Accept, request evidence, or escalate to a senior examiner.

  8. Readiness and the summary written back, only after a pass or the reviewer’s decision.

  9. Which checks ran, what was flagged, who decided.

Examiners spend their time on the decision. Who sent what, which dates disagree and what’s still missing is already worked out.

Result

Written

Case marked ready

  • Case 60418 moved to the decision queue
  • Evidence summary attached for the examiner
  • No reviewer time used in preparation

Work coordinated: input gathering, system lookups, repeated checks and approved updates. Select a step to inspect its role.

Human-owned: process rules, exception judgment and authorization. Synthetic records and transitions; no model calls or live system writes.

Why NodeFox fits

Prepare the evidence before asking for judgment.

Combine document interpretation with explicit completeness checks and record lookups. A reviewer sees the assembled evidence before an approval condition releases the next action; the workflow records each step's output.
Explore the execution platform
What changes

The same work, with the checking done first.

  • Today

    An examiner opens every attachment to work out what the file contains.

    With NodeFox

    Attachments are classified and indexed when the case arrives.

    We measure

    Preparation minutes per case

  • Today

    Missing evidence is noticed mid-review and the file goes back to the queue.

    With NodeFox

    Gaps are flagged before the case reaches an examiner.

    We measure

    Missing-evidence rate and backlog age

  • Today

    Contradictions between documents depend on someone spotting them.

    With NodeFox

    Dates and amounts are cross-checked on every file, with page references.

    We measure

    Conflicts caught before decision versus after

  • Today

    Nobody can say later why a file was treated as complete.

    With NodeFox

    Every readiness decision is attached to the run that made it.

    We measure

    Reopen rate against an audited sample

Raw input → usable output

Prepare better cases while preserving accountable judgment.

A specialist receives a complete evidence bundle and the reviewer’s referral, without mistaking case preparation for a coverage determination.

Input 1
Case CA-411 · equipment damage
Input 2
Policy: PL-290
Input 3
Incident date: Oct 27
Input 4
Repair estimate: $3,850
ILLUSTRATIVE OUTPUTClaims and case processing

CA-411 · specialist referral

  • Evidence bundle complete
  • Reviewer decision recorded
  • Case owner notified in sample flow

Decision attached

Request specialist assessment

Source references and configured checks accompany the result.

Inspect the structured record
{
  "case_id": "CA-411",
  "policy_id": "PL-290",
  "evidence_complete": true,
  "outcome": "specialist_referral",
  "coverage_determined": false
}

Synthetic sample files with inputs, checks and decisions. System updates require configured access and authorization. Documents and reports depend on the chosen Writer format and template.

Illustrative workflow

Prepare better cases while preserving accountable judgment.

See the inputs, checks, human decision and completed output. Sample data; configured integrations.
Illustrative claims and case processing product demonstration
Illustrative workflow · sample data. Product-derived UI; simulated integrations and results.
Transcript and demonstration details

Illustrative execution, not a customer deployment. Narration is AI-generated. Actions and outputs depend on workflow configuration.

  1. 0–8s · Prepare the case before the decision

    An equipment-damage case includes a policy and repair estimate. An inspection photo is initially missing.

    Case CA-411 · equipment damage · Policy: PL-290 · Incident date: Oct 27 · Repair estimate: $3,850

  2. 8–17s · Check completeness and reference dates

    The workflow identifies missing evidence and checks the policy period. When the photo arrives, the review package can proceed.

    Case CA-411 · equipment damage · Policy: PL-290 · Incident date: Oct 27 · Repair estimate: $3,850

  3. 17–27s · Interpret documents inside a controlled process

    AI summarizes the material. Readers retrieve policy context. Code applies completeness rules before routing to a qualified reviewer.

    Case intake · Policy records · Document repository

  4. 27–36s · Keep the determination with the reviewer

    The reviewer requests specialist assessment. The workflow records this referral without making an automated coverage determination.

    Reviewer sees evidence summary · Request specialist assessment · No automated coverage determination

  5. 36–46s · A review-ready specialist referral

    The case output contains the complete evidence bundle, accountable owner and recorded decision. The next reviewer receives the context.

    CA-411 · specialist referral · Evidence bundle complete · Reviewer decision recorded · Case owner notified in sample flow

Make the opportunity measurable

Value the preparation. Preserve the judgment.

Estimate evidence gathering and file preparation separately from the reviewer’s decision. Validate completeness and source quality as well as time; the model does not value unreviewed regulated determinations.

Editable hypothetical example · USD

What could less coordination be worth?

Change four assumptions to estimate annual capacity. These are example inputs, not customer results. Calculations stay in your browser.

Annual capacity potentially released
1,802.7 hours
Annual labor-capacity equivalent
$90,133
Reduction in hands-on time
52%

8,320 cases / year × 13 minutes removed per case ÷ 60. Includes eligibility, exception effort and rework when entered.

Capacity is not automatically cash savings. Value depends on how the released time is used and what the workflow costs to operate.

What it connects to

Your systems stay the system of record.

  • Claims intake inbox or portalreadEmail attachments, portal exports or files in a watched folder.
  • Case system — policy, claim history, statusread + writeThrough its API. Writes are limited to readiness and the summary, and gated.
  • Evidence checklist by claim typereadA maintained table of required documents and tolerances.
  • Team email or chatwriteNotify the reviewer or examiner. Optional.
Who is involved

People keep the decisions that matter.

Case handler

Works the gaps and conflicts that reach them, with the evidence summary already built.

Senior examiner

Takes escalated contradictions and makes the coverage call.

Process owner

Maintains the evidence checklist per claim type and the routing rules.
Where to start

Start with one claim type

Pick the claim type with the most files sent back for missing evidence. We run its real (or anonymised) cases through the checks, measure against your current preparation and decide on production from the results.

Attachment classification and completeness checks for that claim type

Evidence summaries and the reviewer step

Case system read access, plus a gated write path once you approve it

A baseline-versus-pilot scorecard

What the pilot measures
Minutes per case
Hands-on preparation time from receipt to decision-ready, by claim type.
Preparation versus decision effort
Share of examiner time spent assembling the file rather than deciding it.
Missing-evidence rate
Share of cases flagged for a missing document, by document type.
Reopen rate
Cases reopened or sent back after a decision, checked against an audited sample.

Questions operations leaders ask

Our case system already tracks status.

It stays your system of record. NodeFox prepares and checks the file before a decision and writes readiness back through the system’s API.

No two case files look alike.

The model step classifies whatever arrives. Completeness and date checks are exact rules in code your process owner can read and change, and anything they can’t settle goes to a person.

Does it make coverage decisions?

No. It gets the file ready and flags what’s missing or contradictory. Decisions stay with your examiners, and the case record only changes after a pass or a reviewer’s sign-off.

Case files contain sensitive personal information.

Sensitive data handling is reviewed during scoping: which fields are read, where the workflow runs and which model provider sees what. Model requests go to the provider you configure, with your keys.

Which case files bounce back most often?

Bring a case that came back incomplete. We’ll discuss evidence preparation, examiner decisions and the scope of a useful first workflow.