GitHub vs LinkedIn Import: What's Real, What's AI

"Import from LinkedIn" and "Import from GitHub" sit next to each other in the same modal in VeriWorkly, and it would be easy to assume they work the same way under the hood. They don't, and the difference is worth being upfront about rather than letting "import" imply parity between the two.

GitHub import is a real integration

This one is genuine OAuth. When you connect GitHub, VeriWorkly pulls your profile plus up to 30 repositories and deterministically — no AI involved — maps your languages to skills and your repos to project entries. Same input, same output, every time.

• Free accounts can import their own connected GitHub account, once per day.
• Paid accounts can import any public GitHub username.

There's one nuance paid users should know: even though there's no visible per-day limit on importing other usernames, there's a hidden cap of 50 imports/day per account. It exists to protect the shared GitHub API token's rate limit from being exhausted by one heavy user, and it isn't surfaced anywhere in the product today — if you hit it, that's why.

LinkedIn import is AI reading your text

This is the one that surprises people. There's no LinkedIn API integration here — LinkedIn doesn't offer one that would support this use case. Instead, you export your own LinkedIn profile (as text, or as the PDF LinkedIn generates, which gets auto-extracted for you), paste or upload it, and an AI model parses it into structured resume data.

That means the output quality depends on how the model interprets your specific export, not a guaranteed 1:1 field mapping the way GitHub import is. It's a good starting point, and worth reviewing before you trust it as-is.

• Free accounts get one LinkedIn import per month.
• Paid accounts get unlimited imports, with no secondary cap — this path isn't bound to a rate-limited third-party API the way GitHub is.

Same checkbox, different mechanics

Both paths share the same "Replace my Master Profile with this data" option and both always create a brand-new resume document from the result rather than editing one in place. That part is identical. What produces the data underneath it is not — one is a deterministic API pull, the other is a language model doing its best with unstructured text. Knowing which is which should change how much you double-check the result before you rely on it.