
We measure it with you
Our engineers build the first case set with your experts, measure quality and cost on it, and walk your team through the results.
Evaluation Lab runs your agents and workflows on your own cases, scores every answer and compares each new version with the last, so you release only what meets your quality bar.
A different model at each step, run on the same 120 cases three times, so the scores compare.
Balance keeps 94% on your cases at 14¢ a case. Quality adds two points for more than twice the cost.
Creating the ticket needs no model at all, so that step costs nothing.
Before AI does real work, you need three answers: is it good enough for your work, did the last change break anything, and what does each answer cost? Evaluation Lab gives you all three, measured on your company's own cases, for every agent, every workflow and every new version.
Quality is measured on the questions your customers really ask, against the answers your experts expect.
Every new prompt, document or model runs the same cases first, so you see what improved and what broke before a customer does.
It finds the lowest-cost mix of models that still clears your bar, so the price of each case is measured, not guessed.
Your team collects real requests: single questions or whole dialogs, each with the answer you expect or the rules it must follow. Judge models score every answer against them, and the same cases run on every new version.

Offer a refund or a replacement, and ask for a photo of the damage, as returns policy §3 says.
Update the address, cancel the order when asked, and confirm both in writing.
No. Offer the nearest Saturday or Monday slot, with its delivery price.
Not for this order type. List the payment options the store accepts.
Every new version runs the same cases. Evaluation Lab compares it with the last one case by case, and names every case that used to pass and now fails.
Can I pay half now and half on delivery?
Passed in v4 · fails in v5 · judge: etalonA workflow is a chain of steps, and each step needs a different amount of intelligence. Understanding a customer's message takes judgement; filling in a ticket takes none. The strongest model on every step pays the top price for work that a smaller model, or a plain script, does just as well.
Judgement goes to a strong model, routine work to a fast, low-cost one, and fixed rules to a script.
The quality bar you set stays where it is. The price of each case comes down.
New models arrive every few months. Run your cases on one, and you know whether it is better or cheaper for your work before you switch.
You set the quality bar and a budget for the search. Evaluation Lab tests each step on its own cases, runs the finalists end to end several times, and brings back three sets with their quality and cost.
Three steps, six candidates at each: five models and a plain script. Ten steps would make 60 million.
A budget for the search and at most 20 finalists, so it never becomes endless. You see the estimate before it starts.
Every candidate first runs on each step's own cases. It costs little and drops most of them.
On all 120 cases, three runs each, with reworded and reordered cases to catch gains that are only luck.
Quality, Balance and Economy, each with its quality and its cost per case as a range.
Evaluation Lab tries the models you allow at each step of a workflow, on your own cases. At every step it keeps the cheapest choice that holds your quality bar, and a plain script wherever no model is needed.


AI earns trust the way people do: measured on real work, then trusted with more. Models, prompts and knowledge keep changing, and each change can move an answer. So every version is measured on your own cases, for quality and for cost, before it goes live and while it runs.
A version is published only when a full run clears your bar. After release, scheduled runs and sampled answers keep checking it, and every miss becomes a new case.
Every case in the suite, on the version you want to release.
Only a full run on the agent's own model counts. Below your bar, the version waits.
A person publishes it, and the run that cleared the gate stays on record.
Your cases run on a schedule, and samples of real answers are checked against the same bar.
A real conversation that went wrong is saved as a test, so the next version is checked against it.

Our engineers build the first case set with your experts, measure quality and cost on it, and walk your team through the results.

With support and recommendations from our engineers, your team writes cases, runs them and publishes a version when it clears your bar.
Start with the questions your customers really ask. A few dozen real cases are enough to begin, and the set grows as new situations come up.
The people who know what a right answer looks like. Our engineers help them turn real requests into cases: single questions or whole dialogs, each with the answer you expect.
Judges score each answer against the answer you expect: is it right, is it grounded in your sources, is it safe to send. Your experts can open any case and see the answer, the verdict and why.
Your cases are written in the language your customers use, and the answers are scored in it.
In Spaceplace Cloud, on the models you connect. Ask us about running it on your own servers, with local models.
Each run shows its cost estimate before it starts, and the cost of every answer stays on record.

Bring the questions your customers really ask. We'll show you how Evaluation Lab scores the answers.