Recruiting: resumes collected, scored and logged
On schedule, the recruiting AI employee collects resumes from your inbox and hiring channels, reads any layout, pulls out what matters, scores each one against the role and logs it in Lark Base. HR reviews a scored shortlist.
In hiring season, HR can't read them all
From a pile of attachments to a scored shortlist
Collects on schedule AI
Pulls from your hiring inbox and upload link; duplicates are removed.
Reads the resume AI
Reads layout like a person, pulling out name, education, experience and skills — scans and images too.
Matches and scores AI
Works out which role it's for, then scores each dimension with reasons.
Logs and reconciles AI
Writes to Lark Base and checks each batch: received = logged + rejected + exceptions.
HR picks who to meet Person
A notification in the group; HR reviews the shortlist and decides who moves on.
Your current hiring channels
Several roles open at once, resumes arriving by email and upload. The AI employee collects, parses and scores them in batches each day and logs them in Lark Base; HR gets the shortlist in their group chat and maintains the matching rules in a table themselves.Deployed on site · hiring inbox and Lark Base
Questions
Could resumes get lost?
Every batch is reconciled: received must equal logged plus rejected plus exceptions. Any mismatch raises an alert naming the missing ones. Duplicates are removed automatically.
What if the role requirements change?
Scoring criteria and matching rules live in a table HR edits directly — no engineering needed.
Does the AI decide who gets hired?
No. It screens and scores; HR decides who to interview and who to hire.
Hand first-round screening to an AI employee
Connect one hiring inbox and see a scored shortlist the same day.