An experiment: an AI job agent that reads every advert before applying

Description and use

JobAgent’s full‑advert reading and tailored CV approach achieved 31 interviews per 1,000 applications, far exceeding bot performance.

Problem

High-volume job applications generate low interview rates; bots submit thousands of generic applications with ~1 interview per 250 applications.

Outcome

In 13 weeks, 10,483 roles passed filters, 8,585 adverts read, 779 scored fit, 481 tailored CVs, 258 approved applications, 8 interviews, and 3 offers—yielding 31 interviews per 1,000 applications, far above bot benchmarks.

Workflow

• Collect job listings via Chrome and extract job cards
• Read full advert and store
• Score each advert against master profile (0‑100)
• Generate tailored UK CV using uk‑cv‑writer skill
• Write cover letter only when it changes outcome
• Verify CV content against master profile
• Approve roles manually before applying
• Apply to approved roles via job‑apply skill

Evidence

Hiring has turned into a numbers game. UK graduate roles draw a record 140 applications each, up to 290 in the busiest sectors, and in the UK test run described below employers regularly mentioned 300 or more.

Source

JobAgent is a folder of skills: plain Markdown instructions (SKILL.md) that Claude follows, plus small Python helpers.

Source

Four people tested JobAgent on their own searches, in the UK and three other EU countries. The figures below are from the UK run only: one scan every two weeks for 13 weeks.

Source

That is 31 interviews per 1,000 applications, against roughly 1 to 4 for the overnight bots in published accounts.

Source