Composite study

This is a product-class write-up, not a named-client story and not a results press release. Counts below describe the pack we actually write (approach points, tools, layers) — not invented bug totals or conversion lifts.

4
Approach points
5
Tools in the pack
5
Layers covered
3
Challenges documented

Overview

Faithfulness on 50 golden questions looked strong. Staging signup emails were dead, SSO duplicated users, and the copilot sat behind a paywall that did not unlock after payment.

Go/no-go used the eval number because it was the only automated signal.

Industry pattern this is grounded in

LLM evals and product QA answer different questions. A golden-prompt pass is not a test plan. Demos hide fallbacks, permissions, and billing.

Highlights

Typical launch bar: Wait until journeys and evals agree — do not ship on demo scores alone.

Challenges

01

The ML team and the QA team did not share a launch checklist.

02

Payment was tested with a mock that always succeeded.

03

Nobody wanted to be the person who said Wait when the demo was scheduled.

How we tested

01

Pair evals with journey tests: signup, billing, permissions, fallback.

02

Severity on product impact, not on model-score deltas.

03

Ship / wait / block with an owner for residual risk.

04

Regression pack in CI so prompt changes cannot silently break checkout.

Starting point vs what we required

Starting point versus the controls we required
Starting pointWhat we required
The ML team and the QA team did not share a launch checklist.Pair evals with journey tests: signup, billing, permissions, fallback.
Payment was tested with a mock that always succeeded.Severity on product impact, not on model-score deltas.
Nobody wanted to be the person who said Wait when the demo was scheduled.Ship / wait / block with an owner for residual risk.

What we delivered

01

Journeys / UI

  • Playwright — Signup, billing, entitlement
02

Adversarial evals

  • Promptfoo — Small golden + adversarial set
03

API / contract

  • Postman — SSO and role checks
04

LLM traces / evals

  • LangSmith — Prompt regression
05

CI regression

  • GitHub Actions — Both layers on PR

Outcomes

We reported two layers: model behavior and product journeys. Entitlement-after-pay was High. Eval green + broken auth is Wait.

Playwright critical path plus a small Promptfoo pack plus a written gate. The launch owner signed the residual risk list — we did not invent urgency, and we did not hide the auth bugs behind a faithfulness score.

Stack

Testing stack used on this product class
LayerToolWhat we validated
Journeys / UIPlaywrightSignup, billing, entitlement
Adversarial evalsPromptfooSmall golden + adversarial set
API / contractPostmanSSO and role checks
LLM traces / evalsLangSmithPrompt regression
CI regressionGitHub ActionsBoth layers on PR

Same method as the rest of our work: severity-ranked findings, pass rules in writing, and a ship / wait / block call. See the evaluation method and tool comparison.