Todd Stuvé
AI without the jargon

You changed the prompt. Is it better, or does it just feel better?

That one question is the whole reason evals exist. An eval turns "it feels better" into "it scored better." Here is the idea in plain English, no code required.

Grounded in Every's model reviews and the How I AI eval interviews June 2026
Start here

Everyone is writing prompts. Almost nobody is checking whether the prompt worked. That check has a name: an eval. It is an exam you give the AI, and this whole page is that one idea, start to finish.

01The one analogy

An eval is an exam you give the AI.

Forget the jargon. An eval is just an exam. A real exam has three parts you already know: the questions, the answer key, and the grade.

An eval has those same three, plus one more that matters most. You give the exam again, every time something changes. Hold that picture. We are not going to swap it for a different one.

02The build

An eval has four parts.

Build them in order. Each part rests on the one before it. To keep it concrete, here is a real eval reviewers run on every new AI model, the senior engineer eval.

The questions Data

A set of real cases to test.

Here: a real production codebase that works but is a mess inside, the kind someone called "vibe-coded slop."

Taking it Task

What you ask the AI to do with each case.

Here: rewrite that codebase from first principles, the way a senior engineer would.

The grade Score

Mark the answer against an answer key for what good looks like.

Here: two human senior engineers did the same rewrite first. Their work is the answer key, and every model is scored out of 100 against it.

The re-take Repeat

The part people miss. Give the same eval again every time you change the prompt or swap the model.

One grade is a snapshot. Re-running it is what turns a one-time check into an eval.

Skip part four and you are back to a vibe check with extra steps. Keep it, and you have a number you can trust as everything else changes.

03The objection

But isn't this just a test?

Here is the question every engineer asks, and it is the right one. The answer is the cleanest way to see what an eval really is.

A normal software test is deterministic. Same input, same output, every time. 2 + 2 is always 4, so you assert the exact answer and it passes or fails. That is the world QA and engineers already live in.

AI is non-deterministic. Ask it the same thing twice and you can get two different answers, both good. And there is rarely one correct answer to assert against in the first place. The moment you write "assert the output equals X," you have lost, because there are a hundred good versions of X.

A NORMAL TEST Deterministic 2 + 2 4 One input, one answer. Same every time, forever. AN AI MODEL Non-deterministic Write a welcome email Version A Version B Version C All good. All different.
A normal test expects one fixed answer. An AI gives many good ones.

So you stop checking for one exact answer and start grading quality on a scale. That switch, from a pass-or-fail assertion to a graded score, is the line between a test and an eval.

04The contrast

Vibe check vs eval.

This is the whole point, so slow down here. Without an eval, you eyeball two or three answers and ship on a hunch. With one, you get a grade you can trust and repeat.

Before
A vibe check

You eyeball a few answers and ship on a hunch. No record, no re-run, no idea whether the next change quietly made it worse.

After
An eval

You get a grade you can trust. Change something, take the exam again, and watch the number move.

The same eval, every model

One eval. Four grades.

Here is the senior engineer eval in full. The questions are a real production codebase, the kind that works but is a mess inside. The ask is blunt: rewrite it from first principles, the way a senior engineer would. To grade it, two human senior engineers did the exact same rewrite first, and their work becomes the answer key. Every model is scored out of 100 against what the humans built, which puts human level in the 80s and 90s. Then the same eval is re-run on each new model as it ships.

FIRST TO HUMAN LEVEL SCORES 80s TO 90s Fable 5 Human engineers Opus 4.8 GPT 5.5 91 63 62 0 100
Senior engineer eval, scored out of 100. Green is Fable reaching human level, violet is the human range (80s to 90s), gray are the earlier models. Source: Every.
Two weeks earlier, the best model scored 63. Then Fable 5 scored 91, matching a human engineer and the first model to reach that bar. The eval caught the jump the moment it happened.
05Read it honestly

Read the score like a skeptic.

Build an eval hard enough to fail. If every answer passes, you learned nothing. The useful eval is the one some answers flunk, because the failures show you exactly where the AI breaks. For months, every model failed this one. That gap was the most honest thing on the page.

A higher score is not automatically good news. Up just means something changed. Maybe the AI got smarter. Maybe the eval was easy to game. Treat a jump as a reason to go look, not a reason to celebrate.

The concept outlasts the tools. Models change, prompts change, the tool you grade with this year is gone next year. The eval, your written idea of what good looks like, is the part you keep.

If you can't measure it, you're guessing. Evals are how you stop guessing.

Your move this week: write five test cases for the one AI task you rely on most. That is your first eval.