The Nepali Community in Korea: How to Find It, and How to Build One

A Nepali worker arrives in Korea, and within a fortnight has joined five Viber groups. One from the agency, one from the factory floor, one for the district, one for cricket, one that somebody's cousin added them to. Between them those groups contain almost everything a new arrival needs to know. They also contain almost nothing they can rely on, because no one is responsible for any of it, nothing is searchable, and the message that actually mattered — a shift change, a document deadline, a hospital appointment — scrolled past at two in the morning underneath four hundred forwarded photographs.

This is not a Nepali problem. It is the shape every migrant community takes when it grows faster than its channels. But it has a Nepali version, and the way out of it is specific enough to be worth writing down.

This guide covers both halves of the question. If you are looking for the Nepali community in Korea, the first half explains where it actually is and how to get into the parts that are useful. If you are the person organising something — a meet-up, a class, a shop, a student association, a workplace noticeboard — the second half covers how to run it so that it reaches people, earns trust, and does not quietly break Korean law.

The size of a community you cannot see

Korea had 2.65 million foreign residents as of December 2024, the highest figure ever recorded, according to the Ministry of Justice's monthly immigration statistics. Of those, roughly 1.41 million are there to work or study — about 1.11 million foreign workers and 307,000 international students.

Those are large numbers. They are also, from the inside, strangely invisible numbers. You can live in Korea for a year, work alongside twenty other foreign nationals, and still have no idea that there is an association for your nationality in your own city, or that the district office runs a free interpretation service, or that the hospital you have been avoiding has a Nepali-speaking volunteer on Thursdays.

The information exists. The community exists. What usually does not exist is a single place where both are visible at the same time.

Why the app on your phone decides who you can reach

Here is the structural problem, and it is the most useful thing in this guide.

Migrant communities in Korea do not share a messenger. According to DAGACHI's own user research into messaging habits by nationality, the split runs roughly like this: Nepali users gravitate to Viber, Vietnamese users to Zalo, Indonesian and Filipino users to Facebook, Chinese users to WeChat. Korean colleagues, employers and public offices are on KakaoTalk.

That research is the platform's own and is presented in its briefs as a working hypothesis rather than published statistics, so treat the detail as indicative. But the pattern will be familiar to anyone who has tried to organise across nationalities in Korea, and its consequences are concrete:

A Nepali group on Viber is invisible to everyone who is not already in it. It cannot be found by search, it cannot be joined by a link someone sees on a noticeboard, and it cannot be discovered by the Bangladeshi worker on the next line who has exactly the same questions.

Official Korean channels are invisible to you. The district office, the support centre, the employer's HR team and the hospital all publish in Korean on Korean platforms. A notice that was genuinely published is, from your side, a notice that never existed.

Every organiser pays the translation cost separately. The person running the cricket group, the person running the remittance advice group and the person running the Saturday Korean class each translate the same information for the same people, badly, from scratch.

The briefs behind this guide call these three things the deficits of language, channel and structure. The language deficit is having no translator and no budget for one. The channel deficit is having no official place where a message is authoritative. The structure deficit is one or two people carrying the whole thing — a single manager handling members of five nationalities, or a single volunteer answering the same question in four languages until they burn out and stop.

Name them and they stop feeling like personal failures. They are design problems, and design problems have fixes.

Finding the Nepali community that already exists

Before building anything, exhaust what is there. In rough order of reliability:

The embassy. The Nepali Embassy in Seoul is the authoritative source for consular matters, and it is the first place to check when something has gone seriously wrong — a lost passport, a death in the family, a labour dispute that has escalated. See the Nepali Embassy in Seoul.

Your workplace, properly. If you came through the Employment Permit System, your employer has obligations and your EPS paperwork has a paper trail. EPS: working in Korea and the full process guide, EPS Korea: the complete process for Nepali workers, cover what you are entitled to ask for.

City and district foreign resident support centres. Most areas with a significant foreign population have one. They run interpretation, legal advice clinics and document help, usually free. The briefs cite the Busan Foreign Residents Support Centre as an example of an organisation running a multilingual group of exactly this kind. These centres are consistently the most underused resource in the country.

University student associations. If you are studying, the nationality association is the fastest route into a functioning network. The briefs describe the Vietnamese Students Association at Chonnam National University running a nationality-based community of around 85 members — small, specific, and far more useful than a group of five thousand. Nepali student life in Korea covers the wider picture.

Religious and cultural gatherings. Temples, churches and festival committees are where diaspora communities have always organised, and they tend to be the place where older arrivals — the ones who know things — actually are.

Online, including here. The Nepali community board, groups and local meet-ups on this site exist for this, as does find a language partner if what you want is practice rather than paperwork. Broader context sits in the Nepali community in Korea and Nepal–Korea associations and networks.

One honest note about online groups. The useful ones are almost always smaller than you expect and more specific than you expect — a group for Nepali workers in one industrial district beats a group for Nepalis in Korea, every time. Specificity is what makes a group answerable.

If you are the one organising

Everything from here is for the person who has decided to run something. A meet-up, a class, a shop, a student society, a workplace noticeboard, a fan club, an advice service. The mechanics are the same and the mistakes are the same.

What a group actually needs

Three things, in this order.

One place where a message is official. Not three places. The single biggest improvement available to most communities is deciding that one channel is authoritative and that everything important appears there, pinned, in a findable form. Five group chats is not redundancy; it is five places a notice might not be.

A language that reaches everyone it needs to. If your members share a first language, you have one problem fewer. If they do not — and most workplace, district and interest groups do not — you need something better than asking a bilingual member to translate as a favour.

Somebody other than you. Communities die when their organiser gets tired. Delegation is not a growth tactic, it is survival.

Open joining or approval

The first real decision is whether anyone can join or whether you approve each member, and the trade-off is genuine.

Open joining suits anything whose purpose is reach — a shop announcing stock, a festival committee, a public noticeboard. You want the link to spread. You accept that you do not know who is in the room.

Approval suits anything where membership means something: a class cohort, a paid members' group, a workplace group whose notices should not be screenshot outside the company, a group for a specific apartment complex. The briefs recommend approval for internal staff groups for exactly this reason, and it is the right default whenever a group's contents are not meant to travel.

You can run both. A public group that announces, and a private group that organises, is a common and sensible pattern.

Joining questions, and the ones you must never ask

If you use approval, you can ask applicants a question or two. This is useful: ask what district they live in, what shift they work, which course they are on, what they are hoping to get from the group. The answers help you sort people into the right sub-group and spot the applicant who is in the wrong place.

Now the hard rule, which the DAGACHI briefs state in every one of the three documents: do not ask for phone numbers, passport numbers or bank account details. Not in joining questions, not in a pinned post, not in a direct message.

This is worth understanding rather than just obeying. A community organiser who collects passport numbers has created a file that is valuable to criminals, is almost certainly being handled in breach of Korean personal data law, and that they have no lawful basis to hold. Migrant communities are already targeted by document fraud and identity theft. Do not build the database that makes it easy. If you genuinely need to verify someone's identity to run a paid service, that verification belongs with a registered business process, not a chat group.

Pinned posts, and the end of answering the same question forty times

The single highest-leverage habit available to an organiser: write the answer once, pin it, and point at it.

Price list. Class schedule. Pickup times. Meeting point. The rules of the group. What to do if you are sick. Where the nearest clinic with an interpreter is. Each of these gets asked continuously, and each only has to be written once. The briefs make this point from the operator's side — that pinned posts and keyword search are what stop a single person answering the same question separately for every nationality in the room — and it is just as true for a volunteer running a cricket league.

There is a second benefit that matters more than it sounds. A pinned, dated, searchable record of what was announced and when is evidence. For an employer, it is proof that a safety rule was communicated. For a student association, it is proof that a deadline was circulated. For anyone who ends up in a dispute, it is the difference between "we told everyone" and being able to show it.

Hand over real authority, not tasks

The structural deficit is the one that kills groups, and the fix is to stop being the only person who can act.

The pattern that works is to promote your most reliable regulars into co-organisers with genuine, scoped authority — one person who approves new members, one who moderates posts, one who runs a regional or language sub-group. DAGACHI PARTNERS, the platform's management console, was updated in August 2026 specifically around this: inviting additional administrators and delegating permissions individually across group information, joining, members, posts and notices, plus managing several groups from one list and spinning up new sub-groups as the community grows.

Whatever tool you use, the principle holds. Sub-groups by nationality, district, shift or course, each with a trusted member running it, is the only arrangement that scales past a few hundred people without consuming the founder.

Know who is actually in the room

Most organisers are wrong about their own membership. They assume it mirrors the people who talk, and the people who talk are never representative.

If your platform can show you the distribution of members by nationality or language — DAGACHI PARTNERS reports nationality distribution and lets you search members by email, nickname or nationality — look at it before you plan anything. A class you were about to run in Nepali may have a membership that is forty per cent Vietnamese. A shop about to stock one kind of festival item may be serving four festivals. This is the cheapest market research available to a community business, and almost nobody looks.

Translation, and the trap inside it

The reason any of this is newly possible is machine translation that is good enough to publish. DAGACHI's proposition across all three briefs is the same: write a notice once in one language and it reaches members in 30 languages, each reading in their own, with the original text available alongside. The platform reports more than 300,000 app downloads across 120-plus countries, 20 domestic and overseas centres and more than 500 partner MOUs as of October 2026.

Used well, that collapses the language deficit. But there is a trap, and the briefs are right to insist on the feature that avoids it: always keep the original visible.

Machine translation fails in specific, predictable places. Numbers and dates usually survive. Legal and contractual language often does not. Medical instructions, wage calculations, deadlines expressed in relative terms, and anything with a negation in it are all places where a plausible-looking translation can mean the opposite of the original. In a fan community a mistranslation produces an argument. In a workplace group it produces an unpaid overtime claim or an injury.

So: translate freely for announcements, invitations, schedules and general information. For anything contractual, medical or legal, treat the translation as a convenience and the original as the document — and get a human to check it. The platform's own answer for the cases that genuinely need a person is Tong by DAGACHI, its on-site live interpretation service, and the 20 centres that provide interpretation and administrative help. More on how the translation works in AI translation.

If what you actually want is to stop needing translation, that is a different and better project: learning Korean and practical Korean.

Trust, impersonation and the scams that target communities

A community with reach becomes a target. This is the part of the fan-community brief that generalises best, because the fraud patterns it describes are not specific to fandom at all.

The three named problems are impersonation accounts, ticket scalping and merchandise fraud — somebody sets up a convincing copy of an official channel, announces something plausible, and collects money. Replace "concert ticket" with "visa consultancy", "job placement", "remittance at a better rate" or "cheap flight to Kathmandu" and you have the exact shape of the frauds that run through migrant communities in Korea.

The structural defence is a channel that can be verified. DAGACHI issues verification badges to institutional and business accounts — in the briefs these are described for agencies, certified organisations and professional service providers — so that members can tell the real channel from a copy. Examples it gives of verified organisational accounts include a corporate group used by CJGPL and Modeni to post factory payroll notices, a city foreign residents support centre, and the Career Hacker Project, which publishes labour and visa information.

Three practical rules, whichever platform you are on:

If you run a group, make the official channel findable and say plainly, in a pinned post, which channels are yours and which are not. Most impersonation succeeds because members have no reference point.

If you are a member, treat any request for payment that arrives by direct message as suspect, regardless of whose name is on it. Official announcements belong in the official channel, in public, where other members can see them and say "this is not right."

Never pay for a job. This deserves its own line, because it is the most expensive fraud in this community and the most persistent.

The legal boundaries — read this before you post

If you take one section seriously, take this one. The DAGACHI briefs set out operating rules for people running groups, and most of them are not platform policy inventions — they are reflections of Korean law that a community organiser can breach without ever realising it.

Job brokering groups are prohibited. The briefs state this flatly: no groups for recruitment or job placement. Employment agency work in Korea is a licensed activity. Running an unlicensed introduction service — even informally, even for free, even with the best intentions — puts you on the wrong side of employment agency rules, and if money changes hands it is considerably worse. Sharing a public job posting is not the same thing as brokering, but the line is closer than people assume, and "I was only helping" is not a defence. If you want to point members at work, point them at the official route: employer-sponsored work visas and the EPS process.

Do not collect members' personal data. Covered above, and repeated here because it belongs in both places. No passport numbers. No bank account details. No copies of residence cards.

Payments and transactions happen outside the group, and they are your responsibility. The briefs are explicit that the platform does not include selling or payment functions, and that trade and settlement occur through the operator's own existing channels — their shop, their link, their bank account. That means the platform is not standing behind your transactions and no dispute resolution comes with it. If you are selling, you are the seller, with everything that implies.

Licensing, registration and reporting obligations are yours. Stated in the briefs as the operator's responsibility, by industry. Selling food, giving financial or immigration advice, running a transport service, operating a private teaching institute and handling remittances are all regulated activities in Korea with their own registration requirements. A community group does not create an exemption. Nor does being a foreign resident, and the stakes are higher for you than for a Korean national, because a business law problem can become an immigration problem. Immigration offices and overstay and Korean visa types are the starting points for understanding how your status constrains what you may lawfully do.

No prohibited items, no unregistered trading. The obvious one, and the one people drift into by degrees — a group for sharing homecooked food becomes a group selling homecooked food, which is a licensed activity.

A caveat on all of the above, offered in good faith: this reflects the platform's stated operating rules and the general obligations they point at. It is not legal advice, the rules change, and your situation has details this page cannot know. Before you take money from anyone, get advice from someone qualified — a 행정사 (certified administrative agent), a licensed accountant, or the free legal clinic at your local support centre.

How communities actually grow

The briefs set out a three-stage roadmap for a community business. It is worth repeating with one important caveat: the documents explicitly mark the timings as a hypothesis based on general community-business patterns, to be validated against real operating data. Treat it as a sensible shape, not a forecast.

Stage one, roughly the first three months: gather. Free, genuinely useful content — practical tips, how-to information, the answers to the questions people actually ask. Recruit through whatever physical presence you have: the shop counter, a QR code on a noticeboard, your existing social accounts. The goal is the first hundred members and, crucially, finding out what their nationality distribution actually is.

Stage two, months three to six: grow. Open sub-groups by nationality and by interest. Promote your regulars to co-organisers. Start running things on a schedule — regular meet-ups, group purchases, events. Rhythm is what turns an audience into a community.

Stage three, six months and beyond: sustain it. Paid classes, product sales, booked consultations, partnerships. The briefs frame this as members becoming regulars and regulars becoming repeat revenue, which is the honest description of how a community business works.

The order matters and it is not negotiable. Communities that try to monetise in month one have neither the trust nor the numbers, and the attempt usually destroys both. Give first, for longer than feels comfortable.

The briefs put the mechanical setup time at about thirty minutes to open a group, a week to get people joining through a QR code or link, and a fortnight or more before the first notices are landing properly. That is roughly right for any platform, and the thirty minutes is never the hard part.

What this looks like in practice

The groups the briefs cite as live are a useful spread of what the model actually supports:

• a manufacturer using a group to publish payroll notices to a multinational workforce
• a city support centre reaching foreign residents in their own languages
• a university student association running a nationality community of around 85 members
• an information project publishing labour and visa guidance
• certified business accounts using verification to establish that they are who they claim

Note what they have in common. Not one of them is a general-interest group for everybody. Each has a defined membership and a defined job. That is the pattern worth copying.

Two specialist cases

Fan communities

If the community you are building is around music or fandom rather than work, the scale is different and so are the problems.

Citing the Korea Foundation and Ministry of Foreign Affairs survey of global Hallyu status published in March 2024, the briefs count roughly 225 million Hallyu fans worldwide as of December 2023, organised into 1,748 fan clubs across 119 countries, with 68 per cent of those clubs centred on K-pop and 66 per cent of fans in Asia and Oceania — Thailand around 19.5 million, Vietnam around 12.1 million, China around 100 million.

The problem the fan-café brief identifies is specific and recognisable: official announcements are published in Korean, so overseas fans receive them second-hand through unofficial fan translation accounts. The consequence is not just delay. It is that an unofficial channel has become the de facto official one, with no accountability for accuracy, and that fans miss entry windows and ticket deadlines because the translation arrived after the event.

For a Nepali fan living in Korea this cuts the other way, and in your favour: you are inside the country where the events happen. The constraint is almost never distance, it is finding out in time and knowing which announcement is real. Nationality-based fan sub-groups — the briefs describe Vietnamese, Thai, Indonesian and Filipino fan groups with local moderators — are the practical answer, and there is no reason a Nepali one should not exist. Concerts, sport and ticketed events covers how ticketing works in Korea, and the scalping warning above applies with particular force here.

Employers, schools and institutions

If you manage foreign workers, students or performers, the briefs make an argument worth taking seriously even if you never use the product.

The illustration they use is the entertainment industry, and the figures are striking. Drawing on an aggregation of 645 members of the top 100 K-pop groups in 2026, 10.2 per cent hold foreign nationality — Japan 4.2 per cent, China 3.2 per cent, Taiwan 1.4 per cent, Thailand 0.9 per cent — drawn from more than 30 countries, with 13 groups where foreign members are the majority and two with no Korean members at all. Meanwhile company notices, practice room rules, safety instructions and contract explanations are still issued in Korean.

The brief's sharpest line generalises far beyond entertainment: a rule that was not communicated may as well not exist, and a contract that was not understood is the beginning of a dispute. Swap the trainee for a factory worker or an international student and nothing about that changes.

What follows is practical. Onboarding information, schedules, safety rules and administrative deadlines should go out in the languages your people actually read. Residence and reporting deadlines — ARC renewals, visa expiry, address changes — are exactly the kind of thing that should be a scheduled reminder rather than an individual's problem to remember, because the consequence of missing one falls entirely on the worker. Getting your Alien Registration Card in Korea and extending your stay in Korea explain what those deadlines involve. And the delivery record, as noted above, becomes your evidence that the notice went out.

What a group will not do for you

An honest list, because the briefs are sales documents and this is not.

It is not a jobs board, and attempting to make it one is the fastest way into legal trouble. It is not a payment system; money moves through your own arrangements, at your own risk. It is not a substitute for the embassy, the immigration office, a licensed agent or a lawyer — a group is where you find out that you need one of those, not a replacement for them. It will not fix a bad employer, though it is often how people discover that theirs is one. And it is no substitute for learning Korean, which remains the single highest-return investment available to anyone planning to stay.

Translation buys you time and removes an obstacle. It does not remove the advantage of understanding the language around you.

Where to start

If you are joining: pick one specific group rather than five general ones, and find the official channel for anything that matters — your employer's, your district's, your embassy's.

If you are organising: open one group, write three pinned posts (what this is, the rules, the answers to the five questions you are already tired of), and recruit a co-organiser before you recruit your hundredth member.

The DAGACHI app, the platform these briefs describe, is available from download the app, with background in about DAGACHI and the DAGACHI community. Organisations wanting a verified account or a partner arrangement should start at business partnership or partner registration.

Related reading on this site: daily life as a Nepali worker, remittance, cargo and flights and the full guide to sending money from Korea to Nepal, doing business across both markets, job fairs, delegations and festivals, and cultural exchange and differences. If Korea is still ahead of you rather than around you, start with Korean visa documents from Nepal.

Sources, and a disclosure you are entitled to

The platform description, operating rules, group mechanics and growth framework in this guide are drawn from three briefing documents published by DAGACHI and World DAGACHI in October 2026, covering community business, global fan communities and entertainment industry use. More about the platform at worlddagachi.com.

Statistics are attributed as those documents attribute them: foreign resident figures to the Ministry of Justice's immigration and foreign policy monthly statistics for December 2024; worker and student figures to Statistics Korea's immigrant status and employment survey and the Ministry of Education's basic education statistics; Hallyu fan figures to the Korea Foundation and Ministry of Foreign Affairs survey of global Hallyu status published in March 2024; and K-pop membership figures to an aggregation of the top 100 groups compiled by Balance Now Insight and K-Profiles in 2026. Figures for app downloads, supported languages, centres and partner agreements are the platform's own, as published in those briefs.

Where the source documents mark something as a hypothesis rather than a confirmed finding — the messenger habits by nationality, and the three-stage growth timings in particular — this guide says so rather than passing it off as established fact.

The disclosure. DAGACHI Nepal is part of the same group as the DAGACHI platform described here. That is a conflict of interest, and you should read the sections about the product with it in mind. It is also why the legal and limitations sections above are written the way they are: a guide that recommended a group for job placement, or stayed quiet about licensing obligations, would be more useful to the platform and considerably worse for you. Our editorial standards set out how we handle this, and advertise and partner covers commercial relationships.

Nothing here is legal, immigration or financial advice. Rules, figures and platform features change. If something on this page is out of date or wrong, tell us — and if you are about to take money from members of your community, speak to someone qualified first.