The résumé that shows its work.

Anyone can write “led the redesign.” Here you link it to the live page, the repo, the file, and we mark what we checked. A hiring team believes you before they finish the line.

Build yoursSee a real one
8+sources we open and check100%of claims link to the real thing€0to build and share, no card

Verified at the source

We check your work where it already lives.

The repo, the package, the live site, the design file. We open each one and confirm it before we mark a line verified, so a claim on your page always points at the real thing.

Words got cheap this year.

AI writes a flawless bullet in a second, so a page full of them says nothing. What still counts is the thing itself. Yours sits one click behind every line.

Built for Claude

Watch Claude build it from your real work.

Connect Claude over MCP. It reads your repos and files, verifies what it can against the source, and writes the page. No copy-paste, no invented facts.

Connect it

Paste one URL into Claude.

One MCP connection runs the whole product from inside Claude: it reads your repos, writes verified entries, redesigns the page and publishes your link. Also works in ChatGPT, Cursor and Codex.

MCP server URLhttps://nextcv.me/api/mcp

Your bearer token: Settings → Connect your AI, once you sign in.

Run this in your terminal:

claude mcp add nextcv https://nextcv.me/api/mcp --transport http --header "Authorization: Bearer YOUR_TOKEN"

Then just ask Claude to build your CV from your repos.

  1. Open Settings → Connectors → Add custom connector.
  2. Paste the URL above, with header Authorization: Bearer YOUR_TOKEN.
  3. Start a chat and ask Claude to build your CV.
  1. Open Settings → Connectors (developer mode) and add a custom connector.
  2. Paste the URL above, with header Authorization: Bearer YOUR_TOKEN.
  3. Start a chat and ask it to build your CV.

Add to Cursor's MCP settings:

{ "nextcv": { "url": "https://nextcv.me/api/mcp", "headers": { "Authorization": "Bearer YOUR_TOKEN" } } }

Add to ~/.codex/config.toml:

[mcp_servers.nextcv] url = "https://nextcv.me/api/mcp" http_headers = { Authorization = "Bearer YOUR_TOKEN" }

Who it's for

Made for whatever you make.

A designer links the shipped screen. An engineer links the repo. Same move, your craft.

01
Engineers↗
Shipped a Postgres MCP server
github + npm12k installs / wk
02
Designers↗
Rebuilt a checkout flow
figma + live23% less drop-off
03
Marketers↗
Ran a launch campaign
live page + data3.2x signups
04
Sales↗
Built an outbound engine
pipeline data$480k pipeline
05
HR & Recruiters↗
Rebuilt the hiring process
process + data41% faster hires
06
Students↗
Built a study-planner app
live + repoused by 400 peers

For teams

Hire the ones who actually build.

NextCV is a talent index you search by verified work: what someone shipped, where it runs, and who vouched for it.

  1. Plug NextCV into your own AI. The same MCP server, with a team key. Claude, ChatGPT, Cursor: whatever your team already uses.
  2. Ask in plain language.“Find engineers who shipped a verified MCP server” and your AI searches the index mid-conversation.
  3. Get proof, not keywords. Every match links to the repo, the live page and the attestations behind it. Save shortlists in chat, review them in the dashboard.
How it works for teams

How it works

From what you did to a page that proves it.

01

Add what you made

Drop in a project, a design, a shipped feature. Or let your AI pull them from your repos.

about 2 minutes
02

Attach the proof

A repo, a live link, a file. We check it against the source and stamp what holds up.

checks run for you
03

Send one link

Share it, or let companies find you by what you verified.

one URL, always current

Put the proof on the page.

Build yoursSee pricing

Free to build and share, no card needed. Pro, with the AI built in, is €29/month.