Loss-run processing

Step 01

Define and Align

Upload sample data, define your output schema, and provide feedback on the hardest cases.

Output schema

PaidNumber · optional: Amount paid, excluding outstanding reserves.
StatusChoose one: Claim status as recorded in the source.
open
closed
reopened
unknown
Claim countNumber: Count of distinct claims after deduplicating updates.
Outstanding reserveNumber · optional: Stated reserve remaining on the claim; not provided when absent.

app.sutro.sh/functions/loss-run-processing

Collecting feedback · 1/3

Structure synthetic loss-run records with correct totals and status.

Example 1 of 3

Paid versus incurred

Model input

Document excerpt

Loss run

Synthetic loss: paid $20,000; outstanding reserve $5,000; incurred $25,000.

Model output

Paid

Status

Claim count

Outstanding reserve

Why? Optional guidance

Step 02

Model Sweep

Compare quality and cost on your examples, then choose a model or use Sutro’s default.

Compare models on expert agreement and cost. The interactive chart uses illustrative measurements.

Illustrative sweep · select Value or Best to use that model in your function.

Step 03

Deployment

Call Sutro Functions like any other software. Swap the model as soon as something more performant is released.

Selected modelWaiting for sweep results.

Real-time

import sutro as so

input_data = {
    "text": "Synthetic loss: paid $20,000; outstanding reserve $5,000; incurred $25,000."
}

result = so.run_function(
    "loss-run-processing",
    input_data,
)

curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/functions/loss-run-processing/run" \
  -H "Authorization: Key $SUTRO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": {
    "text": "Synthetic loss: paid $20,000; outstanding reserve $5,000; incurred $25,000."
  }
}'

Batch

import sutro as so

job_id = so.batch_run_function(
    name="loss-run-processing",
    data="loss_run_processing.csv",
)
results = so.await_job_completion(job_id)

curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/batch-inference" \
  -H "Authorization: Key $SUTRO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "loss-run-processing",
  "inputs": [
    {
      "text": "Synthetic loss: paid $20,000; outstanding reserve $5,000; incurred $25,000."
    }
  ]
}'