掘金 (Juejin)
Platform · this deep dive is built from the official pages of the platform itself and of the tools that support it.
Overview
掘金 (Juejin; the name translates roughly as "to mine for gold") is a Chinese platform for technical publications aimed at developers: articles, short notes, columns and tag-based collections.
The honest frame for this article is set at once: the platform's substantive pages are closed behind bot protection. Every address — the home page, an article, the user agreement — returned, when we checked, a stub page reading "Please wait…" with a verification script; none of the platform's own text is in it. Everything you read below is taken from the machine-readable files the platform serves openly: robots.txt and its sitemaps.
From those files the structure of the site can be reconstructed with confidence: articles, short notes, columns, tags, collections, author profiles and a news section. The robots.txt also shows what is not visible from the outside: the sections /creator (the author dashboard), /editor, /drafts, /books/payment and /book/order — that is, the platform has both an author dashboard and paid material.
Who's here
The platform does not publicly report audience figures — or if it does, we were unable to read them, for the reason above. We are not substituting any outside estimates here.
The only thing that can be measured is the size of the sitemaps. This is our own measurement of the platform's files as of the check date, not a claim by the platform: the index of article sitemaps lists 54 nested files, and the first of them contains 49,999 article addresses. The author index lists two nested files. The order of magnitude is visible from this, but the exact number of publications and authors cannot be obtained this way: sitemaps need not include everything.
The language is Chinese: this follows both from the name and from the make-up of the tags (see below). A discrepancy with the guide's database matters here: in it the interface language for this platform is recorded as "en". The attribute lang="en" is indeed present in the markup served by juejin.cn — but that is the markup of the bot-check stub page, not of the platform itself. The discrepancy is recorded as a separate entry in the facts file.
Getting started
The registration procedure, the account requirements, the presence of a business profile and of verification were beyond our ability to check: the user agreement page is closed behind the same stub. We will assert nothing about requirements for a phone number or documents.
Indirectly, the robots.txt shows that the platform has personal settings (/user/settings), drafts (/drafts, /news-drafts), an editor (/editor), sign-in through third-party services (/oauth-result) and an author dashboard (/creator). All of these addresses are closed to search robots — the usual practice for personal sections.
What you can publish
The formats can be reconstructed from the sitemaps the platform lists in robots.txt: articles (posts), short notes — 沸点 (pin), columns (columns), collections (collections), tags (tag), news (news), and also separate maps of a given author's publications and notes (user_posts, user_pins).
The subject matter is programming. The tag sitemap lists 725 addresses, among them Angular.js, React.js, Django, Node.js, Ruby and Go; tag addresses take the form /tag/<name>?sort=newest, that is, the list under a tag has a sort order and "newest" is one of the options.
Limits on length, on file size and the list of prohibited content cannot be read from outside: the platform's rules are closed behind the stub.
How to grow
The platform did not let us read the ranking mechanics, how recommendations are built or the rules on self-promotion. The only verifiable indication of how display is arranged is the sort parameter ?sort=newest in the tag addresses from the tag sitemap: since there is an explicit "newest" option, there are other sort options too, but exactly which ones is unknown.
We neither confirm nor deny the existence of an ad dashboard.
Path to monetization
Whether the platform pays authors we were unable to check. There is an indirect sign that paid material exists: robots.txt closes the addresses /book/order and /books/payment to robots — that is, ordering and payment for courses or books exist on the platform. Who that money goes to, and whether authors take part in sharing it, does not follow from the open files.
Tools and automation
We found no official programmatic interface with documentation. The guide's database held an entry for this platform about a "developer section" at the address api.juejin.cn, found on the home page; it had already been marked as corrected (an address with the word api in it is not the same thing as documentation). Our check confirms this: the address api.juejin.cn returns a short JSON linking the site and the mobile apps (applinks), not a description of methods.
The old domain juejin.im redirects to juejin.cn when checked.
Limits and rules
- Bot protection. Substantive pages are served only after the verification script has run. The stub at juejin.cn loads scripts from the domains
byted-static.comandibytedapm.com— this is an observation about the infrastructure in use, and we draw no conclusions about the platform's owner from it. - What is closed to crawling. robots.txt bars robots from
/search,/s/,/editor,/drafts,/user/settings,/creator,/notification,/translate,/zhuanlan,/spost,/appview,/book/order,/books/payment,/subscribe/subscribed,/equationand/oauth-result. - What is open. The same file lists nine sitemaps — for articles, authors, news, columns, tags, notes, collections and authors' publications.
- Legal. We were unable to read the user agreement, so we write nothing about an age threshold, jurisdiction or grounds for a block.
Who it's for
A cautious conclusion. The platform may be useful if your goal is a Chinese audience of developers and you have an author who writes in Chinese: the make-up of the tags shows the usual set of web and server development topics, and the size of the sitemaps speaks of a large body of publications.
It does not suit you if you expect to work with the platform remotely and programmatically: we found no documented interface, and the site is closed behind a bot check, which rules out both external analytics and automation.
It does not suit planning that requires precise rules: the body of rules on self-promotion, the limits and the terms of monetization are inaccessible to us, and working on a platform whose rules you have not read is risky.
Verified data
The checked data this deep dive rests on.
About the platform
the platform describes itself like this: «掘金是面向全球中文开发者的技术内容分享与交流平台。我们通过技术文章、沸点、课程、直播等产品和服务，打造一个激发开发者创作灵感，激励开发者沉淀分享，陪伴开发者成长的综合类技术社区。»
Self-description: this is the platform's claim about itself, not an independent assessment
source, checked 2026-07-28
the platform's substantive pages are closed behind a bot check: the home page, an article page and the user agreement return a «Please wait…» stub with a verification script; none of the platform's text is in the markup
We do not bypass the protection. The stub loads scripts from the domains byted-static.com and ibytedapm.com — an observation about infrastructure, from which no conclusion about the owner is drawn
source, checked 2026-07-28
the old domain juejin.im redirects to juejin.cn
Checked along the redirect chain
source, checked 2026-07-28
What you can publish
from robots.txt and the sitemaps the platform has: articles (posts), short notes 沸点 (pin), columns (columns), collections (collections), tags (tag), news (news), author profiles (user) and separate maps of an author's publications and notes (user_posts, user_pins)
The same file bars robots from /creator, /editor, /drafts, /news-drafts, /user/settings, /notification, /translate, /zhuanlan, /spost, /appview, /search, /s/, /equation, /oauth-result, /subscribe/subscribed
source, checked 2026-07-28
How the money works here
robots.txt closes the addresses /book/order and /books/payment — ordering and payment for courses or books exist on the platform
source, checked 2026-07-28
API access
the address api.juejin.cn returns a short JSON linking the site and the mobile apps (applinks), not a description of methods; no documentation of a programmatic interface was found on the open pages
Confirms correction an address containing the word api is not developer documentation
source, checked 2026-07-28
the address https://api.juejin.cn was found on the home page by the «api/dev» marker in the address; the page is served by a script and is empty without a browser, so whether it is documentation is not verified
CORRECTED the previous version declared any address containing the word api or dev found on the home page to be a developer section. All 57 such records were checked by opening the address: 24 turned out to be documentation. See data/technical/dev-links-verified.tsv and docs/research/S195-догадка-по-адресу.md
source, checked 2026-07-28
Languages
the interface language of the home page: en
The language is taken from the markup attribute, the versions from hreflang
source, checked 2026-07-28
Sources diverge: the attribute lang="en" in the markup served at juejin.cn belongs to the bot-check stub page, not to the platform's interface
DISCREPANCY WITH THE DATABASE: the entry «home page interface language: en» (confirmed) rests on the markup of the protection page. The platform is Chinese; the interface language needs rechecking by a method we cannot apply
source, checked 2026-07-28
Catalogue section: all similar See also: catalogue index · region: China · scheduler comparison · find by situation · platform restrictions