Verified builder profiles for the AI-coding era

Show how you actually build with AI — and get hired for it.

vibexp reads your own Claude, Codex, and Cursor sessions — right on your machine — and turns them into a verified builder profile that shows how you architect, steer, debug, and ship with AI. Then companies on MinuteWork hire you to build for them.

Runs in your own coding agent, on your own subscription. Takes ~15–30 min. You'll get a link to your profile when it's done.

Open source. Your code, prompts, and transcripts never leave your machine — only your finished profile does. Read exactly what's sent →

See example profiles

@ada-vibes

Systems Architect

Verified

Steering 9.4Execution 9.1Engineering 9.2Product 8.6Planning 9.1

Sessions analyzed: 18
Episodes: 142
Scored episodes: 96
Total prompts: 34,870
Tokens written: 443M
Projects: 28

Signature moves

Decomposes ambiguous goals into crisp, reviewable steps
Steers the agent with precise, falsifiable acceptance criteria
Catches regressions before they land

Evidence highlightsredacted

Architecture

Asked the agent to keep the read path and the write path separate so each could be reasoned about on its own.

Evidence line a1b2c3d4

Planning

Turned a vague request into four numbered steps and confirmed the first before moving on.

Evidence line c3d4e5f6

Steering

Set the edge cases first, then asked the agent to implement only after confirming the acceptance checks.

Evidence line 29304152

This is what building well with AI looks like.

Real profiles from real builders. Click any card.

@ada-vibes

Systems Architect

prompts: 34,870
tokens: 443M

34,870 prompts · 443M tokens written

Consistently frames work as a system: clarifies the boundary, names the invariants, then drives the agent through small, verifiable steps.

@rapid-rae

Rapid Prototyper

prompts: 34,870
tokens: 443M

34,870 prompts · 443M tokens written

Consistently frames work as a system: clarifies the boundary, names the invariants, then drives the agent through small, verifiable steps.

@quinn-builds

Product-Minded Builder

prompts: 34,870
tokens: 443M

34,870 prompts · 443M tokens written

Consistently frames work as a system: clarifies the boundary, names the invariants, then drives the agent through small, verifiable steps.

One command. Your agent does the work.

Run the command

Paste it into your terminal. It launches your own Cursor, Codex, or Claude session — no new tool to learn, no API key.

Your agent analyzes your work

It reads your local sessions, scores them against a fixed open rubric, and cites real evidence for every score. Nothing is uploaded but the result.

Get your profile

A shareable, verified builder profile — and a spot in the MinuteWork network where companies hire vibe engineers.

Not a vibe. A measurement.

Eight dimensions, scored from your actual sessions, each backed by real (redacted) excerpts.

ArchitecturePlanningSteeringDebugging & RigorCode QualityExecutionProduct InstinctCommunication

Evidence excerptredacted

Don't write the code yet. First, list the edge cases, then design the types, then we'll implement.

Evidence lines 35–44

We built this the opposite of a black box.

Open source

Read every line of the script and the skill before you run it.

Local-first

Your working tree, .env files, and raw transcripts stay on your machine.

Your subscription, your agent

The analysis runs in your own Claude, Codex, or Cursor. We don't see your prompts.

Only the profile leaves

You can see the exact payload before it sends.

Hire builders by what they've actually shipped with AI.

MinuteWork companies don't read résumés — they read verified builder profiles. Browse the network, see real evidence, and hire a vibe engineer to build it for you in minutes.

Build with the network

MinuteWork networklive soon

@ada-vibes34,870443M

@rapid-rae34,870443M

@quinn-builds34,870443M

Hiring pipeline

Invited
Interview
Offer
Hired

Your next role is in your terminal history.

Free. Open source. ~15 minutes.