An AI recruiting agent is an autonomous agent that screens applicants against your role criteria, schedules interviews, drafts candidate communications, and keeps your applicant tracking system updated without a human babysitting every step. Fleece AI runs one across Greenhouse, Lever, Ashby, Gmail, Google Calendar, and Slack, with approval gates wherever you want the final say.
At a glance
| Best for | Recruiters, talent teams & hiring managers |
|---|---|
| Key integrations | Greenhouse, Lever, Ashby, Gmail |
| Connects to | 2,500+ apps via managed OAuth |
| Setup | Connect your tools — no code required |
| Autonomy | Suggest-only to fully autonomous, with approval gates |
| Pricing | 7-day free trial, then paid plans |
By Loïc Jané · Updated June 8, 2026
How the agent works
From trigger to action — Fleece AI does the work in between.
Gmail
New application
Candidate sourced
Fleece AI
Fleece AI at work
Google Calendar
Interview scheduled
Gmail
Candidate emailed
Slack
Shortlist posted
Gmail
New application
Candidate sourced
Fleece AI
Fleece AI at work
Google Calendar
Interview scheduled
Gmail
Candidate emailed
Slack
Shortlist posted
Hiring drowns recruiters in coordination, not judgment
The work that fills a recruiter's day is rarely the work that requires a recruiter. Resume triage, calendar tetris, and ATS hygiene crowd out the conversations that actually move a hire forward.
How the recruiting agent works
Connect your stack
Link Greenhouse, Lever, or Ashby alongside Gmail, Google Calendar, and Slack through managed OAuth. For tools without an API, the agent can drive a real browser.
Define the role and rules
Tell the agent your must-haves, screening questions, interview loop, and tone of voice. Set approval gates on anything sensitive, like sending a rejection or moving a stage.
Delegate the pipeline
The agent screens new applicants against your criteria, proposes interview times, drafts personalized outreach and updates, and logs every action back into the ATS.
Review and run on schedule
Approve the actions you flagged, then set a recurring flow so the agent works the pipeline every morning, on triggers, or around the clock.
What the agent handles for you
Consistent screening
Every applicant is read against the same criteria, so qualified people surface fast and judgment is applied evenly.
Hands-off scheduling
The agent reconciles candidate availability with interviewer calendars and books the slot without the email back-and-forth.
Drafted candidate comms
Personalized invites, updates, and rejections are written in your voice and queued for your review before they send.
An ATS that stays current
Stages, notes, statuses, and next steps are logged automatically across Greenhouse, Lever, or Ashby as work happens.
Agent teams for volume
A lead agent can delegate to specialized child agents, one per requisition or region, working independently like a real recruiting team.
You stay in control
Approval gates pause the agent on anything you flag as sensitive, so it never sends or advances without your sign-off when it matters.
Connects to your recruiting stack
Fleece AI connects to Greenhouse, Lever, Ashby, Gmail, Google Calendar, and Slack out of the box, plus 2,500+ other apps through managed integrations. When an app has no API, the agent can use a real browser instead.
What you can expect
Frequently asked questions
No. The agent screens, schedules, drafts, and updates records, but you decide how much autonomy it has. Approval gates let you require sign-off before it sends a message, advances a candidate, or takes any action you consider sensitive.
It connects to Greenhouse, Lever, and Ashby for your ATS, plus Gmail, Google Calendar, and Slack for communication and scheduling. Fleece AI also connects to 2,500+ other apps, and can drive a real browser for tools that lack an API.
You define the must-haves, screening questions, and signals for each role. The agent reads each applicant against those criteria and surfaces the strongest matches, applying your rules consistently across the whole pile rather than scoring on hidden logic.
Yes. Because Fleece AI supports hierarchical agent teams, a lead agent can delegate to specialized child agents, for example one per requisition or region, so multiple pipelines are worked in parallel like a real recruiting team.
No. You can set the agent up as a recurring flow that runs on a schedule, such as every morning, or fires on triggers like a new application. It works the pipeline continuously between your reviews.