AI agent for recruiting
An AI agent for recruiting is a software agent that does a recruiter's desk work. A Monmo agent writes job posts, researches candidates, summarises applications against the criteria you give it, arranges interviews and keeps your applicant tracking system current, asking before anything is sent or changed. Decisions about people stay with people: you interview, choose and make the offer.
Updated 2026-11-13 · Free to start. Paid plans from $15 a month.
The job description, line by line
Each line is a duty that appears in typical job descriptions for this role. Beside it is how the agent handles that line and how far it goes on its own.
- Does itResearch, reading, drafting and building. Nothing leaves your business.
- Does it, asks firstThe last step sends, posts, pays or changes a record, so it shows you a card first.
- Prepares it, you decideThe line needs your judgement. The agent brings the facts and a recommendation.
- Stays with youCalls, meetings in person, signatures and relationships.
| # | In the job description | How the agent does it | How far it goes |
|---|---|---|---|
| 01 | Meet hiring managers to define the role, must-have criteria, salary range and interview stages. | Before the intake meeting it drafts a role brief in Google Docs from similar past roles: duties, must-haves, nice-to-haves, interview stages and the questions still open. You and the hiring manager settle the criteria and the salary range. | Prepares it, you decide |
| 02 | Write job descriptions and post openings on job boards, careers pages and social networks. | Writes the job description from the agreed brief, in plain words with the must-haves first, and a version for each job board and for LinkedIn. Once you approve the text it posts through Greenhouse, or in its own browser. | Does it, asks first |
| 03 | Search LinkedIn, job boards and résumé databases for candidates and send personalised outreach. | Researches people who meet the must-haves on LinkedIn and job boards and lists them in Google Sheets with the evidence for each. A personal note is drafted per candidate. Emails go after your approval; LinkedIn messages you send yourself. | Does it, asks first |
| 04 | Review applications against the criteria and prepare a shortlist for the hiring manager. | Reads every application in Greenhouse against the written criteria you gave it and nothing else. For each must-have it notes met, not met or unclear, quoting the line from the résumé. It rejects no one: you choose the shortlist. | Prepares it, you decide |
| 05 | Run phone and video screening interviews with candidates. | It does not interview anyone. Before each screen it gives you a page in Google Docs: the candidate's background, the criteria still unproven and five questions that would settle them. Your notes afterwards become a scorecard for you to check. | Stays with you |
| 06 | Schedule interviews, coordinate interviewers and collect their feedback and scorecards. | Checks the interviewers' Google Calendars, offers the candidate three time slots and sends the invitations when you allow it. The next day it drafts a Slack reminder to any interviewer whose scorecard is still missing. | Does it, asks first |
| 07 | Tell candidates where they stand at each stage and record notes and status in the applicant tracking system. | As a stage changes it drafts the candidate's update, including a considerate no for those not going forward, and prepares the status and notes in Greenhouse or Lever. One card shows the messages and the record changes before either happens. | Does it, asks first |
| 08 | Order background checks and collect references before an offer is made. | After you name the finalist, it prepares the background check request with your provider and emails each referee your standard questions from Gmail, both on your approval. The replies are gathered into one summary; what they mean is your call. | Does it, asks first |
| 09 | Report on hiring progress: time to fill, pipeline by stage and source of hire. | Every Monday it reads the pipeline in your applicant tracking system and writes a page per open role in Google Docs: candidates at each stage, days open, where applicants came from and what has stalled, with last week's numbers alongside. | Does it |
| 10 | Decide with the hiring manager who moves forward and who receives an offer. | This decision belongs to people. The agent sets the finalists side by side in a Google Doc: evidence against each criterion, the interviewers' scores and open concerns, in the same order for everyone. It gives no verdict and no ranking. | Prepares it, you decide |
| 11 | Make and negotiate job offers and complete new-hire paperwork. | The offer conversation and the signatures are yours. Once terms are agreed it drafts the offer letter from your template, prepares the paperwork for signing in DocuSign and builds the new starter's checklist, each for you to review and send. | Stays with you |
| 12 | Attend job fairs, campus visits and recruiting events. | It cannot be at the stand. Beforehand it writes a one-page summary of each role to hand out and a sign-up sheet in Google Sheets. Afterwards the names you collected get a drafted follow-up each. | Stays with you |
6 of 12 lines handled by the agent, some after your approval
Where the duties come from. The twelve lines are the duties that appear most often in public job descriptions for recruiters and talent acquisition specialists, reworded, and checked against the O*NET task list for human resources specialists. Checked against O*NET OnLine: 13-1071.00 Human Resources Specialists
A week with your AI recruiting agent
Reads it against the written criteria and adds a summary, with the evidence for each must-have, to the role's review list.
Prepares a page for each of today's interviews and checks that every invitation has a room or a video link.
Shows one card: the candidate updates it drafted and the status changes for your applicant tracking system.
Adds ten researched candidates to each open role's sourcing list, with a drafted note for each.
Drafts a reminder to any interviewer whose scorecard is missing.
Reports each open role: candidates by stage, days open and where applicants came from.
What you give it on day one
- Your applicant tracking system, such as Greenhouse or Lever, plus the hiring mailbox and calendar
- For each role: written must-have criteria, the salary range and the interview stages
- The interviewers, their calendars and the stage each one covers
- Your templates: job description, candidate emails, the considerate no and the offer letter
- The rules: what may never be used to assess a candidate, and how long candidate data is kept
What stays with you
- Interviews, whether by phone, video or in person.
- Who moves forward, who is turned down and who gets an offer.
- Offers, negotiation and signatures.
- Job fairs, campus visits and other events.
- Sending LinkedIn messages. It drafts them; automated or bulk messaging is not something it does.
Hire one in three steps
- 01
Write the role
Create an agent, name it, and brief it like a new recruiting coordinator: the roles you are filling, how your process runs, and what may never be used to judge a candidate. Paste your hiring policy if you have one.
- 02
Give it access
Connect your applicant tracking system, Gmail and Google Calendar with your own accounts and choose that this agent may use them. It reads your open roles and past job posts first, to learn your stages and wording.
- 03
Set the first routine
Give it one open role with written criteria and ask for a summary of every application by 9:00 tomorrow. Check its evidence against the résumés, then agree the routine: summaries as applications arrive, a report every Monday.
First things to ask
- Here are the five must-haves for the Customer Success Lead role. Summarise each of the 42 applications against them, with the evidence. Do not rank or reject anyone.
- Find 45 minutes this week when Aiko, Ben and the candidate are all free. Show me the invitation before it goes.
- Draft a kind, specific no for the six candidates we are not taking forward. I will read each one before it is sent.
- Every Monday at 9:00, report each open role: candidates by stage, days open and where applicants came from.
Watch it work
A sample replay with example data, sped up. Pause it, scrub through it, or take the mouse yourself.
On screen
Google Calendar
Slack
Pick a job. Watch it get done.
- Why
- You asked for an hour this week with Kenji, Aiko and Jamie.
- Where
- #launch in the Fieldwork Slack
- What to expect
- Posts one message with the two times. Sending the calendar invite is a separate request.
Your team. Connected to everything.
Google Calendar - Week of 15 February+Sample replays with fictional data, sped up. Pause one, scrub through it, or take the mouse yourself.
Questions people ask
What is an AI agent for recruiting?
An AI agent for recruiting is a software agent that does a recruiter's desk work: writing job posts, researching candidates, summarising applications against written criteria, arranging interviews, updating candidates and reporting on the pipeline. A Monmo agent works in your applicant tracking system, mailbox and calendar, and asks before anything is sent or changed.
Can an AI agent replace a recruiter?
It takes over the writing, research, summaries, scheduling, updates and reports. It does not interview, decide who moves forward, make or negotiate an offer, or meet candidates at events. Those need a person, and candidates expect one. A recruiter using it spends less of the week on calendars and more of it with people.
Does it decide which candidates are rejected?
No. It applies only the written criteria you give it, shows the evidence for each one and does not reject or rank anyone out. A person reads the summaries, decides who moves forward and approves every message. Check the hiring and data protection rules where you recruit before using any software in assessing candidates.
Can it source candidates on LinkedIn?
It researches people on LinkedIn, notes why each fits the must-haves and drafts a personal message. You send the messages yourself, or approve them one at a time. It does not run automated connection requests or bulk messaging, which LinkedIn's terms do not allow. Email outreach is sent from your mailbox after you approve it.
Does it work with Greenhouse, Lever and other applicant tracking systems?
Yes. Greenhouse and Lever are the two it is used with most, and others such as Workable and Ashby are in the app directory. The agent uses the account you connect and has the permissions that account has. Where a connection does not cover a step, it can use the system in its own browser.
How does it schedule interviews?
It checks the interviewers' calendars, asks in Slack when two of them clash, and offers the candidate three time slots by email. Invitations with the room or video link go out after you allow them. If someone cancels, it proposes new times and drafts the note to the candidate.