Make your .cv
Drop your PDF resume and your resume in Markdown.
We pack them into one .cv file along with a clean HTML
rendering and, if you want, semantic embeddings for AI search. Nothing
leaves your browser. No account, no upload.
Don't have a Markdown copy yet? Ask ChatGPT or Claude.
Upload your PDF to ChatGPT or Claude, then paste this prompt:
Convert this resume to clean Markdown that an ATS or LLM can parse perfectly. Keep every section, every role, every date, every bullet. Use # for the name, ## for section headers (Summary, Experience, Education, Skills, etc.), ### for individual roles or schools. Italicise dates with *. No tables. No emojis. No commentary. Output only the Markdown, nothing else.
Save the output as resume.md or paste it into the box
below.
1 · PDF
2 · Markdown
3 · Metadata
Want to add BGE-M3 semantic embeddings? Generate them with the Python
package (pip install "cvfile[embed]"):
run embed(markdown) then pack with
pack(..., embeddings=encode_embeddings(payload)). The model is
~285 MB so it runs once locally on your machine, not on every visitor's
browser. The cv CLI is reader-only (extract, inspect, validate,
search) and does not generate embeddings.