◎ CASE / 008◎ AI AGENT◎ KR◎ 2024◎ SHIPPED
Mid-size IT Co.POWERED BY CREWLESS
HR Resume Screener
HR Resume Screener
ROLE
AI Agent Design & Build
PERIOD
2024
MARKET
KR
HEADLINE
-90% · Initial screening time
(01) Challenge · The problem
“Starting point”
Each job posting brings a flood of applications, yet first-pass review remained a manual read-through by HR staff. Backlogs delayed outreach to strong candidates, and screening criteria drifted subtly from reviewer to reviewer.
(02) Solution · How we solved it
“Approach”
Applications are parsed with Python, and Claude evaluates each candidate's fit against the job requirements. Results and reasoning are organized automatically into a Notion database, so HR starts second-round review from a consistently scored candidate list.
(03) System · How it runs
Running on the engine.
INTAKE
RESUME · DOC
CREW
LESS
LESS
ORCHESTRATOR
PARSE AGENT
RESUME EXTRACT
MATCH AGENT
CRITERIA SCORE
>pipeline · hr-candidate-screener · production▌● IN PRODUCTION
Runtime log · Sample run
▌
(04) Outcomes · Results
Proven by numbers.
-76%
Initial screening time
First-pass review, once a document-by-document read, was drastically shortened by agent pre-screening.
Auto
Requirements matching
Every applicant is evaluated against the same requirements, reducing reviewer-to-reviewer variance.
24/7
On-arrival processing
Screening runs the moment an application arrives, so no review backlog builds up.
(05) Stack · Tools used
Trusted tools only.
01 · AI
- ▸Claude API
02 · Backend & Workspace
- ▸Python
- ▸Notion API
▸ Related · AI AGENTALL PROJECTS →
START / FREE CONSULT · NDA OK