Wizhut Research

Research

Small, real products built to answer a question

Some ideas only give up their answer once you build them and put them in front of people. This is where we do that: working products, free to use, each one testing something we were not sure about — and each one publishing what it found.

1 live experimentMore on the way

A question, not a roadmap

Each experiment starts from something we genuinely do not know the answer to. If the answer were obvious, building it would be product work, not research.

Free while it is an experiment

Experiments are how we learn, so charging for one would be charging you for our education. They stay free, and the ones that graduate into products say so loudly before anything changes.

The evidence ships with it

Every experiment has a page here with its method and its measurements, including the parts that did not go the way we expected. A claim without a number behind it does not go on these pages.

Experiments

What is in the lab right now.

Each one opens onto its own page: the question it is testing, how it works, and every measurement we have taken. The headline figure below is the single most useful thing that experiment has told us so far.

Promptivo

Live experiment

Can prompt quality be measured without asking a model?

A deterministic scorer for AI prompts: seven research-backed merits, a regression calibrated once against an LLM judge and then frozen into coefficients. It runs with no model in the loop, so the same prompt always scores the same and your prompt never leaves the machine.

Free, with an accountNo LLM at scoring time

ρ = 0.53

against a held-out judge consensus — where two frontier judges only manage ρ = 0.61 with each other

Use it How it works, and what it found

The lab is deliberately small

One experiment at a time gets real attention; the rest are notes and prototypes until one of them earns a page here. There are several ideas queued behind the one above — they arrive when there is something working to point at, not when there is a plan to announce.

If an experiment stops being interesting, it is retired and its page stays up with what it found. That is the deal that makes the measurements on these pages worth reading: nothing here is under pressure to look like a success.

Promptivo

The one you can use today.

Promptivo scores a prompt across seven quality dimensions in milliseconds, with no model call and no prompt ever leaving the machine. Free, permanently, with no quota — an account just gives your history somewhere to live.

Try Promptivo Read the method