STEP 01
WHO WE RECRUIT
Source the specialist
Specialists are sourced against the work the queue is producing — not a generic talent pool. The bench grows where the deficits are.
→HUMAN DATA OS · WORKFORCE · ONE BENCH OF REVIEWED SPECIALISTS
The people behind every model.
One contract. One record. One bench of reviewed specialists deployed onto annotation, review, and approvals.
BENCH COVERAGE
ROSTER · QUEUE
Standing by CANDIDATES · Standing by OPEN OFFERS
● LIVE
SPECIALISTS
Standing by
On the bench. Reviewed. Ready for the next queue.
CERTIFIED
Standing by
Cleared on the rubric your team already runs.
OPEN DEFICITS
Standing by
Coverage gaps the bench is recruiting against now.
HOW IT WORKS
One bench. One record.
Source the specialist. Vet them on the rubric. Deploy onto the queue.
STEP 02
WHAT WE REVIEW
Vet the work
Every specialist is cleared on the rubric the team already runs. Trial cases, reviewer notes, and certification records sit on file.
→STEP 03
WHERE THEY LAND
Deploy to the queue
One contract. One payout record. Deployed onto annotation, review, and approvals — under the same rubric the model is graded on.
METHODOLOGY · FIVE STAGES
Queue, match, schedule, close the gap, learn.
The bench runs on one operating loop. The same five stages are the same on every program — annotation, review, approvals, and specialist execution.
01 · QUEUE
Bring work into one managed queue
Annotation, review, approvals, and specialist tasks all enter one system with policy, SLA, and ownership attached. No queues hidden in side channels.
Task batch · rubric · policy rule → Queued work with clear routing rules
02 · MATCH
Assign the right reviewer or specialist
Skills, calibration, quality history, and current load decide who should take the work next. The reason for the match stays on the task.
Expertise tags · reviewer history · capacity → Named owner with a reason
03 · SCHEDULE
Book the shift with pricing attached
Week and month calendars handle drag-drop shift assignment, surge-aware pricing, and conflict detection. The pricing band is on the shift, not in a spreadsheet.
Shift template · surge band · conflict rules → Assigned shift with pricing
04 · CLOSE THE GAP
Fire auto-recruit when the roster is short
Deficit signals flow into the gap matrix. When the gap crosses a threshold, auto-recruit rules fire to source new specialists against the work the queue is producing.
Deficit count · rule triggers · priority → Auto-recruit run + sourcing pipeline
05 · LEARN
Every outcome feeds the next run
Completed tasks update routing memory, staffing signals, QA records, pricing, and SLA configuration. The bench keeps a record of how it gets better.
Decision history · QA feedback · follow-up actions → Better routing and staffing memory
ROSTER · BENCH VIEW · ANONYMIZED
The bench is filterable.
Filter by certification, trust tier, availability window, or program. The roster surfaces specialists you can actually deploy — not a directory of every name on file.
CERT: HIPAA
TIER: CALIBRATED
AVAILABILITY: 48H
PROGRAM: CLINICAL
Names anonymized. Example bench view from the platform workforce directory.
DEFICIT · GAP MATRIX
When the bench is thin, auto-recruit fires.
The deficit matrix is not a chart on a dashboard. It is a live read of the coverage gaps across roles, skills, and shifts — and the trigger that fires auto-recruit when a gap crosses the threshold.
Specialists are sourced against the work the queue is producing — not a generic talent pool. The bench grows where the deficits are.
AUTO-RECRUIT FUNNEL · LAST FIRE
OPENING
Robotics surge
CLEO SCREEN
37 reached
SHORTLIST
8 qualified
ASSIGNMENT
3 booked
GAP MATRIX · ROLES × SKILLS × SHIFTS
SKILL × PGM
EVAL
ANNOTATE
APPROVE
CLINICAL
8
22
48
VISION
0
16
34
ROBOTICS
26
42
12
SAFETY
6
18
30
● LOW
● MEDIUM
● HIGH
TRUST TIMELINE · ONE SPECIALIST
Trust is earned on the record.
A specialist's tier is not a label. It is the result of calibration, review, incidents, and certifications — every change auditable, every change reversible.
W-8 · CALIBRATION
Gold-set pass · 0.91
First calibration on the rubric the team runs. Trial set scored, reviewer notes filed, tier opened.
↳ TIER ACTIVE
W-4 · REVIEW
184 tasks · 96 QA
Active queue load with reviewer agreement above the floor. The bench treats this as a trusted contributor.
↳ TRUSTED
W-2 · INCIDENT
1 span reopen
Reopen on a single span. Coached, not penalised. The reason and the resolution stay on the specialist's record.
↳ COACHED
NOW · TIER CHANGE
Clinical cert active
New certification activated. Specialist now eligible for clinical NLP queues. Audit record updated.
↳ CALIBRATED
WORKFORCE LIVE METRICS
The cockpit reads off the platform.
The tiles below come from the live telemetry feed — not a marketing artifact. The same numbers move the scheduler, the auto-recruit triggers, and the deficit matrix.
ACTIVE WORKERS
Standing by
Reviewed roster with calibration on file
CERTIFIED
Standing by
At least one active certification on file
OPEN DEFICITS
Standing by
Role × skill gaps the deficit dashboard flagged
JOBS FILLED · 30D
Standing by
Offers accepted across Cleo, AI Interviewer, and Workforce
QUEUE
184 tasks
2 SLA risks
COVERAGE
92%
late shift thin
DEFICIT
48h
clinical annotation
CLEO
8 ready
shortlist attached
BENCH READOUT · LAST 7 DAYS
SURFACES · FOUR PILLARS
Four surfaces. One bench.
The Roster shows who's on the bench. Detail tells you how they got there. The Scheduler is where the shift gets booked. Auto-recruit is where the next specialist comes from.
ROSTER
Workforce landing
Real worker roster, filterable by cert, skill, availability, and trust tier — with the deficit chart and a bulk-assign action.
DETAIL
Worker detail
Certification history, earnings, quality trend, active assignments, and the calibration snapshot from the rubric.
SCHEDULER
Week + month view
Week and month calendar with drag-drop shift assignment, surge-aware pricing, and conflict detection on the same surface.
AUTO-RECRUIT
Rules + dry-run
Rules list, rule detail editor, dry-run preview, trigger-now action, execution history, and the enable/disable toggle.
WHAT COMES OUT
What your team leaves with.
The bench is the artifact. Who's on it, what they cleared, what they reviewed, and what they were paid — all on the record.
01
Specialist roster
Every specialist on the bench, the rubrics they're cleared on, and the queues they're deployed to.
↳ ARTIFACT
02
Review records
Every cleared case, reviewer note, and overturned call — attached to the specialist who made it.
↳ ARTIFACT
03
Certification reports
Who passed which rubric, on what date, against which trial set. The record the auditor asks for.
↳ ARTIFACT
04
Payout records
One contract per specialist. One ledger per queue. The money trail matches the work trail.
↳ ARTIFACT
PROGRAM NOTE · ON THE RECORD
“One bench. One rubric. One contract. Coverage and calibration stopped being two different conversations the day the roster and the queue lived in the same record.”
QUESTIONS · BEFORE YOU STAFF
What teams ask before they put the bench on a queue.
Six questions, six answers. The same six come up across annotation, review, approvals, and specialist execution.
01 · ON CONTRACTS
How are specialists contracted?
One contract per specialist, one payout record per queue. The money trail matches the work trail. Every cleared task is auditable against the contract terms it cleared under.
02 · ON CERTIFICATION
Whose rubric clears them?
Yours. Specialists are vetted on the rubric the team already runs. Trial cases, reviewer notes, and certification records sit on file. Re-certification follows the same rubric version.
03 · ON ASSIGNMENT
How are tasks matched to specialists?
Skills, calibration, quality history, and current load. The match comes with the reason — the recruiter or program lead sees why a name surfaced.
04 · ON PRICING
How is pricing surfaced on a shift?
Pricing lives on the shift, not in a side spreadsheet. Surge bands, off-peak bands, and SLA configuration are visible on the same surface as the assignment.
05 · ON DEFICITS
What fires auto-recruit?
Threshold rules on the deficit matrix. When a role × skill × shift gap crosses the threshold, auto-recruit runs against the rule definition with a dry-run preview before it executes.
06 · ON HANDOFF
Where does a cleared specialist go next?
Into the queue they were sourced for. The cleared rubrics, the trust tier, and the audit trail travel with them. The handoff is the record, not a summary.
WHAT CHANGES
What changes when the bench is on one record.
Specialist work breaks when the roster, the queue, the rubric, and the contract live in four different tools. The team chases coverage in one system, calibration in another, scheduling in a third, and payouts in a fourth. By the time a release is up, nobody can answer who cleared what — or why.
On one record, the bench is the artifact. Who's on it, what they cleared, what they reviewed, and what they were paid — all on one ledger. The deficit fires recruiting. Recruiting fires calibration. Calibration fires assignment. And every outcome feeds the next run.
WHERE IT FITS
In the loop, this is where you recruit.
Test the run. Review the hard cases. Recruit the right specialist. Remember the misses. Approve what's right.
01
Test
02
Review
03
Recruit
● YOU ARE HERE
04
Remember
05
Approve
RELATED MODULES
Next to this in the Human Data OS.
CLEO
The chair of record for every review.
Every override, override note, and signed call — attributed and on file.
See the page →ANNOTATION
Annotation that doesn't drift.
Same rubric. Same reviewers. Same standard across every batch.
See the page →EVALUATION STUDIO
Test it before it ships.
Codify the rubric, score every release against it, and ship with proof attached.
See the page →WORKFORCE
The people behind every model.
One bench. One contract. One record. Bring the work your team is already doing — we'll bring the specialists who can do it.