WORKER DETAIL
Worker A.L.
Clinical NLP reviewer
TRUST TIER
Calibrated
ACTIVE
3
COMPLETED
1842
QUALITY
97%
WORKFORCE DEMO · HUMAN DATA OS
See who makes the data.
The data only people can give starts with people you can vouch for. Walk the roster, the worker record, auto-recruit, and the hiring pipeline — every reviewer carries a cert, a trust score, and a record of work you can defend under audit. Read-only here. Hands-on in a pilot.
ACTIVE WORKERS
0
CERTIFIED WORKERS
0
OPEN DEFICITS
0
OVERVIEW CONSOLE
Four surfaces. One question: who did the work.
ROSTER GRID
6 workers
Skills, certs, trust tier, availability, and current assignment.
SCHEDULER
7 day week view
Shift blocks, surge bands, conflicts, and coverage gaps.
AUTO-RECRUIT
Rule-driven outreach
Deficit trigger, rule evaluation, shortlist, assignment.
DEFICIT MATRIX
0 snapshot deficits
Role by shift heatmap with close-the-gap routing.
ROSTER · SIX WORKERS
Filter by cert, skill, trust tier, or availability.
Every worker is identity-verified, certified for the work, and tiered by trust earned over time — so the people behind your data survive an audit. The six rows here are illustrative seed data; in a pilot they read from your live roster.
| Worker | Skills | Certifications | Trust tier | Availability | Active assignment | Quality |
|---|---|---|---|---|---|---|
Worker A.L. Clinical NLP reviewer | NLP · HIPAA · Oncology | HIPAA · Oncology QA | CALIBRATED | Now | Clinical summary eval | 97 |
Worker M.S. Vision QA specialist | Segmentation · Bounding boxes · QA | SOC2 · Segmentation | CALIBRATED | Now | Mask audit surge | 95 |
Worker R.K. Robotics demonstrator | Bimanual · Demonstration · UR5 | Lab access · UR5 | TRUSTED | 2 days | Bimanual package-sort | 91 |
Worker P.H. RLHF preference labeler | RLHF · Preference · Rubric | Policy · Preference | CALIBRATED | Now | Preference batch R-18 | 96 |
Worker T.O. Safety reviewer | Safety · Bias · Red-team | Safety · Bias | TRUSTED | 1 week | Escalation review | 93 |
Worker J.B. Code review specialist | Python · TypeScript · Review | Code QA · TypeScript | CALIBRATED | Now | Tool-call audit | 94 |
WORKER DETAIL · DRILLDOWN
Trust built over time, not claimed once.
Every worker page carries a trust-score timeline so operators see why the tier is what it is — and when to recalibrate.
TRUST-SCORE TIMELINE · 12 WEEKS
W-12
W-10
W-8
W-6
W-4
W-2
W-0
LATEST · 97 · RECALIBRATED EVERY 2 WEEKS
AUTO-RECRUIT · RULES FLOW
Deficit in. Offer out. Evidence everywhere.
A 40-hour gap on the skill-by-shift matrix becomes governed outreach: a rule fires, a shortlist is scored, and the dispatch carries the deficit and the rule that triggered it. Every step is recorded, so the fill is auditable, not ad hoc.
STEP 1
Trigger
Deficit ≥ 40 hours detected on the skill × shift matrix.
STEP 2
Evaluate rule
Match the deficit against active auto-recruit rules.
STEP 3
Dry-run preview
Recruiter sees the candidate pool and expected reach.
STEP 4
Dispatch
Preview the Cleo outreach route with the job attached.
STEP 5
Close deficit
Accepted candidates fill the shift; rule marks closed.
CLEO RECRUITING PASS
● 40H GAP
94
Candidate N.K.
Cleo interview route
● REVIEW
88
Candidate M.P.
Assignment-ready
● BOOK
82
Candidate R.S.
Human review
● HOLD
DEFICIT MATRIX
Role by shift, in hours.
The numbers are open hours. The biggest cells are the deficits auto-recruit will route first.
| Early | Mid | Late | |
|---|---|---|---|
| Reviewer | 0 | 12 | 48 |
| Annotator | 6 | 0 | 22 |
| Specialist | 18 | 4 | 34 |
DEFICIT READING
PIPELINE · KANBAN
Sourced to hired, on one board.
SOURCED
3 CARDS
Candidate N.K.
Candidate M.P.
Candidate R.S.
SCREENED
2 CARDS
Candidate T.O.
Candidate A.L.
IN INTERVIEW
1 CARD
Candidate P.H.
OFFER OUT
1 CARD
Candidate J.B.
HIRED
1 CARD
Candidate M.S.
NEXT STEP
Run it on your own roster.
This walkthrough is read-only. Start a pilot and the same roster, trust scores, and hiring pipeline read from your live metrics — with the people behind your data verified and on record.