Resume & CV Parser: PDF to Markdown & JSON
Pricing
from $10.00 / 1,000 resumes
Resume & CV Parser: PDF to Markdown & JSON
Parse PDF resumes and CVs into clean, LLM-ready Markdown that keeps every section: contact, summary, experience, education and skills. Text pages are read directly and scanned pages get OCR. For recruiters, ATS imports and AI candidate screening in n8n, Make or Zapier. Failed files are free.
Pricing
from $10.00 / 1,000 resumes
Rating
0.0
(0)
Developer
PANIX AI LAB
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Categories
Features
Share
Turn PDF resumes and CVs into clean Markdown with every section intact, ready for your ATS, a spreadsheet or an LLM that screens candidates.
- Sections kept in order: contact details, summary, experience, education, skills and certifications.
- Scanned CVs work too: text pages are read directly and scanned pages with OCR.
- No OpenAI key, and candidate data stays private: the models run inside the Actor.
Try it now: the input is already filled in (one sample CV). Click Start and the results appear in a table you can export to Excel, CSV or JSON. That example run costs about 2 cents, which the $5 of free monthly credit on Apify's free plan covers.
Price: from $0.01 per resume, and you pay only for results you get. The full price table is in the Pricing section below.
Example result
One result, shortened. Every field is described in the Output section below.
{"source_url": "https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf","status": "success","candidate_markdown": "## Jordan Ellis\n\n## Senior Data Analyst\n\nAustin, TX, USA | jordan.ellis@example.com | ...\n\n## EXPERIENCE\n...","processed_at": "2026-11-01T18:09:47.000000+00:00"}
What you get
Turn PDF resumes and CVs into clean, LLM-ready Markdown that keeps the candidate's sections intact: contact details, summary, experience, education, skills and certifications. Pages that already contain text are read directly, and scanned pages are read with OCR. Layout analysis (IBM Docling) and OCR (RapidOCR) run inside the Actor with their models built in, so you need no OpenAI key and no candidate data is sent to a third-party AI service.
Paste resume links, click Start, and get one JSON result per resume with the full text as Markdown. You pay only for resumes that parse.
What does it do?
For each resume URL you provide, the Actor:
- Downloads the PDF.
- Checks each page: pages that already contain text are read directly, and only scanned pages go through OCR.
- Reads the layout: columns, headings, lists and tables, in the correct reading order.
- Exports the whole resume as Markdown, with a heading for each section and tables as Markdown tables.
- Saves one result per resume to the dataset, with a clear
successorfailedstatus.
Who uses it
- Recruitment agencies: turn every incoming CV into clean text for search, shortlists and AI summaries.
- ATS and HR-tech builders: a parsing step for uploaded resumes, called through the API.
- HR automation in n8n, Make and Zapier: parse resume attachments from a mailbox and fill Google Sheets, Airtable or your ATS.
- AI screening and agents: give an LLM the resume and the job description and get a match score or summary, through the API or MCP.
- Talent search teams: load the Markdown into Pinecone, Qdrant or Supabase and search candidates by skills and experience.
Why this parser
- Reads the layout, not just the text: columns, sidebars, headings, lists and skill tables stay in reading order. Two-column templates come out column by column instead of mixed line by line.
- OCR only where it is needed: pages with a text layer take a few seconds; scanned pages are detected and read with OCR.
- Private by design: all models run inside the Actor. No OpenAI key, no external OCR service, nothing sent to a third-party AI.
- Fast and cheap at scale: about 5 seconds per resume after start-up. $13.50 per 1,000 resumes on the Starter plan, less on Scale and Business.
- Fair billing: broken links, web pages, non-PDF files and unreadable PDFs are free, and the run stops at your maximum charge.
- Safe with any link: only public addresses are fetched, signed URLs work, and tokens in links never appear in the log.
Coming from another resume parser?
| Problem people often run into | How this Actor handles it |
|---|---|
| Two-column templates come out mixed line by line | Layout analysis keeps each column and sidebar together, in reading order |
| Scanned resumes come back empty | Pages without a text layer are detected and read with OCR |
| Every link costs money, even broken ones | Failed downloads, HTML pages and unreadable files are free |
| Resumes are sent to an outside AI service | Parsing runs inside the Actor; nothing leaves it except your results |
| Fixed JSON fields that don't match your ATS | Markdown with every section, ready for your own schema in one LLM step |
| A big batch goes over budget | The run stops before the resume that would exceed your maximum charge |
How to use it
- Click Try for free and open the Input tab.
- Paste direct links to PDF resumes into Resume / CV PDF URLs, one per line. The prefilled link is a sample CV for a first test.
- Click Start. A run with one resume takes about 20 seconds.
- Open the Output tab: Overview shows the status of each resume, Markdown shows the parsed text.
- Download the results as JSON, CSV, Excel or HTML, or connect them to your tools (see API and integrations below).
Input
| Field | Type | Description |
|---|---|---|
documentUrls | array of strings (required) | Direct public links to PDF resumes or CVs, up to 50 MB each. Each link must return the file itself, not an HTML preview page. |
One resume (the sample CV with fictional details):
{"documentUrls": ["https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf"]}
A batch from your own storage (signed links work):
{"documentUrls": ["https://your-bucket.s3.amazonaws.com/cvs/candidate-1042.pdf?X-Amz-Signature=...","https://www.dropbox.com/scl/fi/abc123/candidate-1043.pdf?dl=1","https://example.com/careers/uploads/candidate-1044.pdf"]}
Files in Google Drive or Dropbox need a direct-download link (for Dropbox, end the link with ?dl=1).
Output
One dataset item per resume. A real result for the sample CV (Markdown shortened):
{"source_url": "https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf","status": "success","candidate_markdown": "## Jordan Ellis\n\n## Senior Data Analyst\n\nAustin, TX, USA | jordan.ellis@example.com | +1 555 010 0199 | jordanellis.example.com\n\n## SUMMARY\n\n...","processed_at": "2026-11-01T18:09:47.000000+00:00"}
candidate_markdown for the sample CV starts like this:
## Jordan Ellis## Senior Data AnalystAustin, TX, USA | jordan.ellis@example.com | +1 555 010 0199 | jordanellis.example.com## SUMMARYData analyst with 7 years of experience turning sales, marketing and operations data into dashboards, ...## EXPERIENCE## Senior Data Analyst, Northwind Outdoor SupplyMar 2022 - PresentAustin, TX- Built the first self-serve sales dashboard in Power BI, used every week by 60 store managers.- Designed a Python demand forecast that reduced stock-outs on the top 100 products by 18%....## SKILLS| Area | Tools and methods ||----------------------|-------------------------------------------------------|| Programming | SQL, Python (pandas, scikit-learn), R || BI and visualisation | Power BI, Tableau, Looker Studio || Data platforms | Snowflake, BigQuery, dbt, Airflow |
A resume that cannot be read comes back with "status": "failed" and an error_message, for example "Download failed with HTTP 404." or "The URL returned an HTML page, not a file. Use a direct download link to the PDF.". You are not charged for it.
Pricing
Pay per event: you pay for resumes that parse successfully, plus a small fee when a run starts.
| Apify plan | Per resume | Per 1,000 resumes |
|---|---|---|
| Free | $0.015 | $15 |
| Starter | $0.0135 | $13.50 |
| Scale | $0.0125 | $12.50 |
| Business and higher | $0.01 | $10 |
- Run start: $0.001 per GB of memory, so $0.004 per run at the default 4 GB.
- Never charged: failed downloads, links that return a web page, non-PDF files, and password-protected or unreadable PDFs.
- Examples (Starter): 1 resume = $0.0175. 100 resumes in one run = $1.354.
- Set a maximum cost per run in Run options: the run stops before the resume that would go over it and keeps everything parsed so far.
Good to know
- Languages: pages with a text layer work in any language. OCR on scanned pages reads English and other Latin-script languages, accents included.
- Speed: about 5 seconds per text-based resume after start-up; a scanned page takes about 40 seconds.
- Files: PDF only, up to 50 MB. Word documents and images come back as
failed(free); save them as PDF first. Password-protected PDFs can't be read. - Links must be public and return the file itself. Links to private or local network addresses are blocked.
- Output format: Markdown with the resume's own sections, not fixed JSON fields (see the FAQ for turning it into fields).
- Memory: the default 4 GB gives 1 CPU. More memory adds CPU and speeds up OCR, and the run start fee grows with it.
API and integrations
Python (pip install apify-client)
from apify_client import ApifyClientclient = ApifyClient("<YOUR_APIFY_TOKEN>")run = client.actor("panix/resume-cv-json-parser").call(run_input={"documentUrls": ["https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf"],})for item in client.dataset(run.default_dataset_id).iterate_items():if item["status"] == "success":print(item["candidate_markdown"][:300])else:print("failed:", item["error_message"])
JavaScript / Node.js (npm install apify-client)
import { ApifyClient } from 'apify-client';const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });const run = await client.actor('panix/resume-cv-json-parser').call({documentUrls: ['https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf'],});const { items } = await client.dataset(run.defaultDatasetId).listItems();console.log(items[0].status, items[0].candidate_markdown?.slice(0, 300));
cURL (runs and returns the results in one call, for runs up to 5 minutes)
curl -X POST "https://api.apify.com/v2/acts/panix~resume-cv-json-parser/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \-H "Content-Type: application/json" \-d '{"documentUrls": ["https://api.apify.com/v2/key-value-stores/gZUDoFSN6sExqUTEm/records/sample-cv.pdf"]}'
No-code tools
- n8n: in the Apify node, run the Actor
panix/resume-cv-json-parserwithdocumentUrls, then read the run's dataset items and mapcandidate_markdowninto the next step. - Make: Apify app → Run an Actor (wait until it finishes) → Get Dataset Items.
- Zapier: Apify app → Run Actor action, then fetch the dataset items of that run.
- Google Sheets, Slack, webhooks: under the Actor's Integrations tab.
AI agents (MCP): add https://mcp.apify.com/?actors=panix/resume-cv-json-parser to Claude, Cursor or any MCP client and ask, for example, "Parse this CV and list the candidate's last two jobs: ".
Data and privacy
- Files are downloaded only from the links you provide and processed inside your run. Nothing is sent to an outside AI or OCR service.
- Results stay in the run's dataset in your Apify account. The log shows only the host and path of each link, never query strings, so signed-URL tokens stay private.
- Resumes contain personal data. Make sure you have a lawful basis to process them (for example under the GDPR), and delete datasets you no longer need.
FAQ
Does it return fields like name, email, skills and jobs as JSON?
No. It returns the whole resume as Markdown with a heading per section, which LLMs read well. For fixed fields, send candidate_markdown to an LLM with your schema, for example: "Extract name, email, phone, current title, skills and jobs (company, title, start, end) as JSON."
Does it work with scanned resumes? Yes. Pages without a text layer are read with OCR. A scanned page takes about 40 seconds, so scanned batches need a longer run timeout.
Can I parse Word (.docx) resumes or photos?
Not yet. Save them as PDF first. Links to other file types come back as failed and are free.
How do I parse resumes from email attachments, Google Drive or Dropbox?
The Actor needs a direct download link. Use Dropbox links ending in ?dl=1, Google Drive direct-download links, or signed S3, Azure or Google Cloud Storage URLs. In n8n or Make, upload the attachment to storage first and pass its link.
How many resumes can one run handle? As many as you like, at about 5 seconds per text-based resume. For large batches, give the run a longer timeout. One broken file never stops the others.
Support
Missing a feature or found a resume that is not parsed well? Open an issue on the Issues tab with a sample link and the run ID, and I will look into it.