Compose a specialist agent for every phase of recruiting.
One agent sources candidates, another screens against your rubric, another coordinates interviews, another re-engages past candidates on signal — every agent rubric-loaded, every screening decision auditable, your recruiters handle the relationship work.
Best fit for · Recruiting teams running structured screening across multiple roles, channels, and a long-tail of past candidatesvs Lindy →Five concrete tasks.
- 01
Specialist agents for every phase of recruiting.
Compose the pack you need: a Sourcer for continuous candidate research, a Screener for rubric-based first pass, a Coordinator for interview logistics, a Re-engagement specialist for past candidates. Add or drop specialists as your roles + channels evolve.
- 02
Rubric loads into every agent.
A rubric custodian (or your rubric memory) holds the role definition: must-haves, nice-to-haves, exclusion criteria, calibration candidates from past hires. Every other agent — Sourcer, Screener, Coordinator, Re-engagement — loads it before acting. Bad-fit candidates drop out before any outreach goes out.
- 03
Source against the actual rubric.
Sourcer pulls candidates matching the role's rubric, not the job description's wishlist. GitHub for engineering, LinkedIn + niche boards for everything else, including off-platform research for senior roles. Heartbeat cadence per role.
- 04
Screen with structured rigor.
Screener applies the same first-pass rubric every time. Output is per-criterion — no hallucinated 'great culture fit' takes. Hiring manager sees a structured comparison, not vibes. Audit log captures every screening decision.
- 05
Re-engage past candidates on signal.
Re-engagement watches past-candidate memory plus public signals — layoffs, promotions, quiet LinkedIn updates. When last year's silver-medalist becomes likely-actionable, the pack reaches out. Persistent memory is the part that's hardest to do manually.
Setup once, then watch it run.
Concrete operator setup, the phases the pack moves through, and where you stay in the loop.
Build ICP
Sourcer turns the role brief into a candidate ICP, hard skills, signals, anti-signals, plus the search-string variants that find them.
Source
Researcher runs the search across LinkedIn and your sourcing tool, dedups against your ATS, and assembles a ranked candidate list.
Outreach
Outreach writer crafts personal first-touch messages anchored in something real about each candidate. The hiring manager approves the batch.
Schedule
Replies route to the Scheduler, which proposes Calendar slots that respect both sides' working hours and writes the screen onto your ATS.
Your pack, your workflow.
Workflows are markdown that reference the roles in your pack. Below is one example shape - yours can have different agents, different steps, different cadence.
A senior sourcer's pipeline volume — composable, rubric-loaded, with structured screening hiring managers trust and a re-engagement loop persistent memory makes possible.
Things teams actually ask.
Run this pack on your team's work.
Closed-beta cohorts are small. Tell us about your work and we'll configure the pack for what you actually do.
Request beta access