Read & learn
Written guides.
Understand Jev, one idea at a time. Walkthroughs, recipes, and writeups — our notes first, the original next.
Explore the Jev ecosystem.
Learn, build, and play with Jev — from deep technical guides to creative experiments.
Watch & learn
Video guides.
See an idea take shape. Tutorials, demos, and deep dives, organized by topic and credited to their creators.
Make & explore
Experiments.
See what builders made with Jev. Games, tools, repositories, articles, and more — traced to their sources.
The directory03 / 03 · Experiments
Find something worth exploring.Experiments
Every category, from games and tools to repositories and writeups, traced to a primary source.
Real-time research powered by
Same ideas.
Different paths.
This is the full registry, not just the featured picks. Figures like speed, cost, and stars are maker-reported or captured snapshots, not JevMade measurements.
19 experiments
Left-pad a string by asking Jev how many spaces it needs. Yes, padStart already exists.
SOURCE SCREENSHOTEXPAND ↗
Make a non-generative decision model talk by holding a 20,000-word tournament for every broken sentence it posts to Discord.
SOURCE SCREENSHOTEXPAND ↗
A joke package that asks Jev whether a number is even, making one network round trip where a remainder would do.
SOURCE SCREENSHOTEXPAND ↗
choice
Is this number even? · true / false
A parody of slow-thinking AI: Bev gets Jev's answer, then delays it with an elaborate paperwork routine.
SOURCE SCREENSHOTEXPAND ↗
from the source
[0:01] Bev has taken a number for your ticket. It is number 4,512.Decision made: billing. Bev is stamping it.Verdict BILLINGBev is 97% confident. (Jev’s actual confidence: 88%)Bev thought for 9s. Actual thinking time: 0.38s.
Chat with a bot that can answer with only five Turkish phrases, while Jev chooses the words and punctuation energy.
SOURCE SCREENSHOTEXPAND ↗
A Magic 8-Ball with twenty custom replies, from “Yes. Stop asking.” to “You already know. You just wanted a witness.”
ACTUAL RECORDING12 SEC ↗
choice
You are a super intelligent and witty magic 8-ball. Choose the best answer to reply to the user’s question. · Yes. Stop asking. / Ask a human with a l...
A pull-request Magic 8-Ball that turns nine GitHub signals into one of four light-review verdicts without approving anything.
SOURCE SCREENSHOTEXPAND ↗
choice
The question is whether this pull request can be approved on a light read, without studying the code. · yes / lean_yes / unclear / no
The People v. Your Drawing is a browser sketch game that challenges you to illustrate an assigned prompt so an automated judge can pick it out of four options.
ACTUAL RECORDING12 SEC ↗
A parody type guard that asks Jev whether a value is a record—and makes another call for the negative form.
SOURCE SCREENSHOTEXPAND ↗
noul
Is `value` a record? · non-null object that is not an array / null, array, primitive, or otherwise not a record
A parody MCP oracle that hashes your words into arbitrary confident answers, but sends serious health, money and safety questions to humans.
SOURCE SCREENSHOTEXPAND ↗
from the source
// Jef. The entire model. A System Zero Model by TypoSafe AI (typesafe.lol).// Jef never writes a sentence. Jef only picks from what you gave it, scores it,function hash(s){ var h=2166136261>>>0; for(var i=0;i<s.length;i++){ h^=s.charCodeAt(i); h=Math.imul(h,16777619)>>>0; } return h; }
Builds an intentionally bad language model by asking Jev which word comes next, one Choice request at a time.
SOURCE SCREENSHOTEXPAND ↗
choice
The text below is an excerpt from a longer document, cut off mid-sentence. Which word or punctuation mark comes next, immediately after the last word...
A pixel-art 8-Ball that scores nine yes-or-no propositions in parallel instead of asking one multiple-choice question.
SOURCE SCREENSHOTEXPAND ↗
noul
Is the user’s text in `question` asking a question that can naturally be answered yes or no? · yes or no
A Victorian 8-Ball whose answer, paradox check and risk score animate its needle, steam and glowing fluid.
SOURCE SCREENSHOTEXPAND ↗
choice
Determine if the statement or question is true/affirmative, false/negative, or ambiguous, and select the corresponding canonical Magic 8-Ball answer....
Scores LinkedIn posts for corporate performance, then invites Grok to roast only the ones that cross the line.
SOURCE SCREENSHOTEXPAND ↗
score
How much of this post is performance rather than substance? · plain direct writing / mostly substantive / noticeably inflated / mostly performance / p...
A voice-controlled claw machine with Jev cast as the fruit-fly brain behind it.
SOURCE SCREENSHOTEXPAND ↗
choice
Which claw command does `call` give? The call can be phrased freely, colloquially, rudely, in English or German, and may address the fly by name. · le...
Ask Jeeves, with Jev behind the search box and a choice of yes/no, multiple-choice or scored answers.
SOURCE SCREENSHOTEXPAND ↗
noul
`instructions` is the visitor’s own question, verbatim, with an empty `state` unless they add context · Yes / No · Choice · Score
Pitch a startup idea to a retro infomercial and compare how stupid Jev thinks it is with how fundable it looks.
SOURCE SCREENSHOTEXPAND ↗
FizzBuzz implemented by asking Jev which term each number gets, one typed Choice per number.
SOURCE SCREENSHOTEXPAND ↗
from the source
const { answers } = await client.systemOne({ state: { number: value }, questions: { term
Check an X draft before publishing to judge whether the maker's girlfriend is likely to be angry about it.
SOURCE SCREENSHOTEXPAND ↗
Jev, briefly
A decision model, not a chatbot.
A chat model generates text token by token and hopes you parse it. Jev never generates a word. Your code sends the state of the world plus typed questions; Jev returns every answer in one parallel pass — typed values with calibrated probabilities and a confidence score your code can trust. About 70–500 ms end to end, $0.042 per million input tokens, output free. Trained with what TypeSafe calls Reinforcement Learning for Calibrated Decisions.
choice Pick one of these.
You define the options. Jev returns the chosen option, a probability for every option, and a confidence. The workhorse of routing, agents, and games — the answer space is always legal, so there is nothing to parse and nothing to hallucinate.
score Rate this on a rubric.
You describe ordered levels — say trivial / normal / critical. Jev returns the level, the probability of each, and a confidence. Scores turn fuzzy judgment (“how severe is this log line?”) into a number-free decision your code can branch on.
noul Is this statement true?
You assert a statement; Jev returns the probability that it is true — a noul. Moderation, verification, guardrails, “does this diff actually fix the bug?”: one question, one calibrated probability.
Official material lives at typesafe.ai and docs.typesafe.ai. JevMade is an independent community registry — not affiliated with or endorsed by TypeSafe AI.
Recurring lessons
Patterns that keep showing up.
One decision per tick
Games, drones, trading bots, and browser agents all converge on the same loop: serialize the state, ask one decisive question, act, repeat. Jev’s latency makes the loop feel instant — the model lives inside the control loop, not outside it.
Confidence as a gate
The answer says what; the confidence says whether to act. The most reliable entries threshold on confidence to route edge cases to a slower model or a human — automation with an honest escape hatch.
Shrink the choice space in code
Makers rarely ask Jev to pick from everything. Local tactics prune 225 gomoku moves to ~40; DOM filters turn a page into an element table; code narrows options, Jev judges within them.
Jev judges, LLMs talk
A recurring split: Jev makes every decision cheaply and instantly, and a small LLM is only invoked when a human-facing string must actually be written. Decision and generation are separate budgets.
Made something with Jev?
Send the link — repo, live demo, post, or video — plus a line on what it does and which primitives it uses. Every entry is verified against its primary source before it ships.
Submit your experimentSign in first so we can reply to you. Submissions are reviewed, not published automatically.