Core API

Feedback & refunds

Thumbs-down any call you don't like and we refund it in full — no ticket, no explanation, no argument. The same feedback is the optional signal behind the opt-in Reliability Loop (Pro).

Why feedback matters

The first reason is the quality guarantee: a thumbs-down on a call id refunds that call automatically, no questions asked. You never argue an invoice line — you flag it and it’s free.

Feedback is also the optional signal behind the Reliability Loop (Pro). If you opt in, the calls you rate and correct become candidates for per-operation tuning — reviewed by your own engineers, in place. This is opt-in and off by default; rating a call never trains a language model on your content.

Posting feedback

Feedback can be posted from server code or the console.

typescriptawait client.feedback.create({
  call_id: "chatcmpl-abc123",
  rating: "thumbs_down",
  correction: "The summary should be one sentence, not three.",
  tags: ["length:over", "accuracy:partial"],
});

bashcurl https://api.newmen.ai/v1/feedback \
  -H "Authorization: Bearer $NEWMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "call_id": "chatcmpl-abc123",
    "rating": "thumbs_down",
    "correction": "The summary should be one sentence, not three.",
    "tags": ["length:over", "accuracy:partial"]
  }'

At minimum, include call_id and one of rating, correction, or tags. You may post multiple feedback rows per call; the latest entry per user wins when building a dataset.

Tagging conventions

Tags are free-form strings, but the team has converged on a dimension:value convention. Examples:

accuracy:good, accuracy:partial, accuracy:wrong
tone:formal, tone:terse
pii:redact-missed, pii:over-redacted

Consistent tagging makes evaluator binding and dataset filtering dramatically easier. We do not enforce a schema; the console will suggest tags you have used before.