AI Recruiter Crew
A crew of agents that screens, calls, and schedules. Recruiters approve.
- Client
- INSNAPSYS (our own product)
- Sector
- Recruitment and HR technology
- Our role
- Product owner, built and operated in-house
- Status
- Running in production
The brief
What was breaking
Recruitment breaks at the coordination layer, not the judgment layer. A recruiter's day fills up with parsing CVs, chasing candidates for a first call, rescheduling that call twice, and typing notes into an ATS. The decisions that actually need a human take minutes. Everything around them takes days.
- Application volume outruns the bandwidth to screen it consistently, so good candidates are missed and screening quality drifts by the hour.
- First-stage screening calls follow the same script every time, but still need a recruiter on the phone for each one.
- Interview scheduling burns hours per role, and every reschedule restarts the loop between candidate, interviewer, and hiring manager.
- Notes, scores, and outcomes get written up after the fact, inconsistently, which makes the pipeline impossible to measure.
What we built
The system we delivered
Application intake and CV scoring
Applications are parsed on arrival and scored against the job description, so recruiters open a ranked shortlist instead of a raw inbox.
AI-led screening calls
A screening agent runs the structured first-stage conversation, records and transcribes it, and scores the candidate against the criteria agreed for the role.
Scheduling that handles the exceptions
Interviews are booked against live interviewer availability with calendar integration, including reschedule handling, confirmations, and meeting links.
Structured recommendations, with reasoning
Every hire/pass recommendation arrives with the evidence behind it, so a recruiter reviews a prepared case rather than reconstructing one.
Human checkpoints where they matter
Agents run the coordination. Advancing, rejecting, and offering stay with the recruiter, and every automated action is logged against the candidate record.
Built with
The stack behind it
The lead tier is what makes this system what it is. Everything under it is the platform that carries it.
AI and agents
Automation
Platform
What changed
The result in production
Time-to-shortlist cut from 14 days to under 24 hours
The pipeline stops waiting on human availability for steps that never needed human judgment.
Consistent screening at any volume
Every candidate gets the same structured questions and the same scoring rubric, whether the role gets 30 applications or 300.
Recruiters spend their time on decisions
Coordination and data entry move to the agents. The judgment calls, and accountability for them, stay with the team.
Same problem?
Let's scope what this would look like for you
Start with a two-week Discovery Sprint. We map your highest-value workflows and deliver a prioritised pilot roadmap grounded in what we have already shipped.