Observe.
Measure.
Conclude.

Proprietary methodology. Observation-based. Fabrication-resistant.

Structured Observation Proprietary Inference Calibrated Deployment Profile

Agents evaluate candidates based on experience and intuition.

Agents observe and record. The system evaluates.

METHODOLOGY FRAMEWORK

Input. Engine. Output.

01 FRAGMENTATION Unverified candidate pool ? ? BSS 02 OBSERVATION 4-channel structured assessment INFERENCE 03 STRUCTURED PROFILE Calibrated dimensional output D1 D2 D3 D4 D5
01

Fragmentation

Unverified candidate pool — self-reported skills, unknown competence, coached presentations.

02

Observation

6-8 hour structured observation across four independent channels. Statistical inference separates coached from genuine.

03

Structured Profile

Calibrated deployment profile with quantified confidence levels and evidence-documented recommendation.

Observe

Trade-specific scenarios under real workplace conditions. Multiple independent observation methods operating simultaneously. 6-8 hours of structured observation. Agents record behaviour through structured protocols.

Infer

Proprietary inference layer. Observation data processed without human judgment. Evaluation separated from observation by architecture. Assessor-independent. Fabrication-resistant by design.

Deliver

Dimensional deployment profiles with quantified confidence per dimension. Context-specific deployment recommendations. Independent fabrication assessment. Every output traceable to observed behaviour.

METHODOLOGY SPECIFICATION

What the Methodology Delivers

Property What the methodology does What this means for clients
Evidence Directly observed behaviour under trade-specific conditions. Task sequences replicating workplace constraints — specification changes, time pressure, quality/speed trade-offs. Measures actual performance under realistic pressure. Not interview skills.
Scope Each protocol maps to a specific ISCO-08 occupational code and regulatory jurisdiction. Assessment conditions derived from destination requirements. Assessment mirrors deployment conditions exactly. Generic protocols do not exist.
Fabrication Multiple independent observation methods operating simultaneously. Coached behaviour cannot sustain across all methods. Detection is architectural, not bolted on. Coached candidates identified before deployment, not after. Cannot be circumvented through preparation.
Output Multi-dimensional profiles with quantified confidence per dimension. Recommendations derived from profiles, specific to trade-country-project context. Nuanced decisions. Strong for standard conditions, flagged for complex environments. System communicates what it knows and what it doesn't.
Duration Adaptive — continues until statistically sufficient confidence across all dimensions. No fixed time. Most complete in 6-8 hours. Consistent confidence levels every time. Additional observation when a profile is genuinely ambiguous.
CLIENT DELIVERABLES

What Clients Receive

DELIVERABLE 01 Deployment Profile
D1
D2
D3
D4
D5
Multi-dimensional. Confidence-quantified per dimension. Context-specific to trade-country-project.
DELIVERABLE 02 Fabrication Assessment
CLEAR
No inconsistency detected across observation methods.
FLAGGED
Inconsistency detected between observation methods. Notation provided.
DELIVERABLE 03 Deployment Recommendation
DEPLOY Carpenter — Structural
Germany · HWK · BG BAU Recommended for deployment in target context.
CONDITIONAL D3 below threshold
Additional observation recommended Same candidate, different context = different recommendation.
6-8 hrs Assessment duration
1308 Trade-country protocols
19 Professions
32 Countries

Related screening topics

Topical referenceCross-border compliance pillarBuyer segmentListed parent & PE portfolio buyersReference librarySkills standards rubric libraryInsightPre-deployment certification — workers who arrive ready

Request a Methodology Briefing

Full methodology documentation available under NDA for qualified enterprise clients.

Request Briefing