AI Services Market
AI Services Market › Use cases › Work & Operations
Work & OperationsDone with Claude Jun 1, 2026

Rank 200 resumes into a shortlist with Claude

A friend uploaded 200 resumes to Claude for a ranked shortlist; the top pick bombed the first call because it ranked writing quality, not job readiness. · Jun 1, 2026

Rank 200 resumes into a shortlist with Claude preview
What They Did
A solo recruiter at a small startup received more than 200 applications for one role and, short on time, uploaded the resumes into Claude to generate a ranked shortlist. She took Claude’s top-ranked candidate into a first-round call, but the post says he could not explain much of what was on his resume; commenters mainly argue this reflected a weak prompt and overreliance on the model.
How It Went
The immediate result was a bad top pick: the #1 candidate looked strong on paper but failed the first screening call. The post’s broader claim is that a general-purpose LLM favored polished, keyword-heavy writing over real job readiness, though it gives no measured shortlist stats beyond that anecdote.
Worth Knowing
The source’s clearest catch is process risk: the recruiter used Claude because manually reviewing 200 resumes felt too slow, but the thread repeatedly flags privacy, audit, and prompting issues.
Try it yourself with Claude

I'm uploading [number] resumes for a [job title] role. Rank them for job readiness, not writing polish, using these criteria: [must-have skills, years of experience, relevant achievements]. Give me a top [number] shortlist with evidence from each resume and flag where you're unsure. Then suggest three screening questions to test the top candidates, since the ranking shouldn't be the final call.

Read the original ↗

Do this with a skill.

These AI Services Market playbooks do this exact job.