Translate resume to English: the steps, the tools, the traps
Published
To translate a resume to English, run the whole document through a translator that keeps the layout, then fix the four things machine translation gets wrong on a CV: job titles, degrees, dates and names. Names of people, companies and schools stay exactly as they are. Everything you describe in your own words (titles, duties, skills, the field you studied) gets translated. The job is an hour of careful checking for one CV, and that hour is worth spending, because a recruiter reads a mistranslated job title as a wrong one.
There are two very different people searching for this. One is a candidate with a CV in Spanish, Indonesian or German who is applying for a job in English. The other is a recruiter, a job board or a developer building an applicant tracking system who keeps receiving CVs in languages the team does not read. This guide covers both, in that order. We build ResumeJSON, a CV parsing API that can return a CV's fields in English, so read the second half as written by an interested party. The first half needs nothing from us.
What to translate and what to leave alone
Most bad CV translations fail in the same places. Before you touch a tool, know which parts of the page change and which do not.
| Part of the CV | Translate it? | What to watch |
|---|---|---|
| Your name | No | Keep the spelling on your passport and your LinkedIn profile |
| Company names | No | "Bank Mandiri" stays "Bank Mandiri", even when the name contains ordinary words |
| School and university names | Usually no | Keep the official name; add an English gloss in brackets only if it helps |
| Job titles | Yes | Use the title an English-speaking employer uses for the same role, not a word-for-word rendering |
| Duties and achievements | Yes | Rewrite into short English bullet points that start with a verb |
| Degrees | Yes, with care | Name the degree, then the closest English equivalent if one exists |
| Skills | Yes | Tool and product names stay as they are |
| Dates | Reformat | Month names translate; the numbers do not |
| Email, phone, links | No | Add the country code to the phone number |
The trap is the middle of that table. A translator that renders every word will turn a company name into a sentence and a job title into something no English speaker has ever written on a CV.
How to translate a resume to English, step by step
This is the manual route, and for one CV it is the right one.
- Start from the editable file. Use the Word or Google Docs original, not a PDF. A PDF can be translated, but fixing the result is slower because the text is harder to edit afterwards.
- Run a first pass through a document translator. The two free routes most people already have are listed in the next section. Both give you an English copy with your layout mostly intact.
- Put your name and every company and school name back. Search the translated file for each one and restore the original spelling wherever the translator changed it.
- Fix every job title by hand. Look up two or three English job adverts for the same role and use the title they use. "Staf Administrasi" is an "Administrative Assistant" or an "Office Administrator", depending on the job, and only you know which.
- Rewrite the bullet points. Machine-translated duties tend to be long and passive. Cut each one to a verb and a result: "Reduced invoice processing time" beats "Was responsible for the process of invoices".
- Handle degrees and grades. Write the degree as awarded, then a plain English equivalent: "Sarjana Teknik (Bachelor of Engineering)". Leave out a grade conversion unless you know the official one, since a guessed conversion can undersell you.
- Normalise dates. Pick one format and use it everywhere: "March 2022 to Present" or "03/2022 to present". Month names are the part a translator most often leaves half done.
- Read it aloud, then ask a fluent reader. A native or near-native English speaker will spot the awkward phrases in five minutes. That check is worth more than any tool.
If an employer, a university or an immigration office asks for a certified or sworn translation, none of the above is enough. That is a legal document produced by a qualified translator, and the organisation asking will say what it accepts.
Free tools that keep your layout
Two tools cover most people, and both translate the whole file rather than a pasted paragraph. As of September 2026:
- Google Docs. Google's own help page describes it in four steps: open the document, click Tools and then Translate document in the top menu, name the copy and pick a language. It creates a new, translated document and leaves the original alone. If the command is missing, the file is still a Word file being edited in Docs, and Google's help page says to convert it to a Google Docs file first.
- DeepL. DeepL's document translation handles Word files, PowerPoint decks and PDFs. Its own page describes the feature this way:
"Drag and drop different documents for easy translation, while preserving original formatting and layout"
| Google Docs | DeepL document translation | |
|---|---|---|
| Input | A Google Docs file (convert Word first) | Word, PowerPoint or PDF |
| Output | A new Google Docs file | A translated copy of the file |
| Layout | Kept in the copy | Kept, per DeepL's page |
| What you still do | Steps 3 to 8 above | Steps 3 to 8 above |
Neither tool knows what a job title means in your industry, and neither knows which words in "Bank Central Asia" are a name. That is why the checking steps exist.
When the manual way is enough
For a candidate translating their own CV, the manual route above is the whole answer. You translate one document, once, and then keep the English version up to date by hand. A parsing API adds nothing to that job: it returns data rather than a finished CV, and you would still have to lay the result out yourself.
The manual way is also enough for a hiring team that receives a foreign-language CV now and then. Open it in Google Docs, translate a copy, read it. Done.
It stops being enough when the CVs arrive in volume and in many languages, and when what you need at the end is data in your own system rather than a document someone reads.
Translating CVs you receive: straight into English fields
A recruiter or a job board with applicants from several countries has a different problem. Nobody needs a nicely formatted English CV. The team needs to know, in English, who this person is, where they worked, what they did and what they studied, stored in the same fields as every other candidate so that search and filters work.
Translating the document and then typing the result into your system does the job twice. The faster route is to translate during extraction: read the CV in its own language and write the fields out in English in one step.
That is what the output_language option on ResumeJSON does. Send a CV in any language with ?output_language=English and the text values come back in English. The rules follow the table at the top of this article:
| Translated into English | Left exactly as the CV writes it |
|---|---|
| Job titles and the headline | Names of people |
| Highlights (the bullet points) | Company and school names |
| Skills | Emails, phone numbers and URLs |
| Degrees and fields of study | Dates, which are normalised to YYYY-MM instead |
| Locations, languages and proficiency wording |
The response says what it understood: meta.outputLanguage reads English, so your code can tell a translated record from one returned in its original language. Leave the option out and the CV comes back in its own language, which is the default.
A request looks like this:
curl -X POST 'https://resumejson-resume-cv-parser-api.p.rapidapi.com/v1/parse?output_language=English' \
-H 'x-rapidapi-key: YOUR_KEY' \
-H 'x-rapidapi-host: resumejson-resume-cv-parser-api.p.rapidapi.com' \
-F 'file=@cv-indonesian.pdf'The answer has the same shape as any other parse: basics, work, education, skills, certifications and languages, with every role carrying start_date, end_date and is_current. The full list of fields is in the reference, and extract information from a resume walks through each one.
A few details matter for a multilingual pipeline:
- The option accepts a name or a code.
English,en,Frenchorfrall work, and other languages work the same way. - Scans work too. A photographed or scanned CV is read as an image, so a paper CV in another language is handled the same way. OCR resume parser covers that route.
- Names are never translated. A candidate stays searchable by the name on their passport, and a company stays matchable against the name your client uses.
- Nothing is invented. A field the CV does not state comes back as
null, in any language.
For a backlog of CVs in mixed languages, the same request runs in a loop; bulk resume parsing covers concurrency and retries, and resume to Excel turns the output into one spreadsheet your team can read.
Try it on one CV first
You do not need a key to see whether this fits. The free resume parser has an Output language box: drop in a CV in any language, type English, and read the fields that come back. It needs no signup, the document is not stored, and it is limited to six parses a minute and 8 MB per file, which is plenty for a handful of real CVs.
When it fits, the paid API is on RapidAPI. The Basic plan is 100 parses a month for $0 with a hard cap, and the paid plans start at $29 a month for 1,000 parses; every plan is on the pricing table.
Which route is yours
- You are translating your own CV: use Google Docs or DeepL for the first pass, then do the checking steps by hand. That hour is the part that gets you the interview.
- You receive the odd foreign-language CV: translate a copy in Google Docs and read it.
- You receive CVs in several languages and store them in a system: translate during extraction, so every candidate lands in the same English fields. Try it free on one of your CVs.