Turn role criteria, job descriptions, skill evidence, and occupational data into structured, explainable decision support through APIs and governed workflows.
Decision-ready workforce intelligence signals
AvelinLabs structures workforce signals so applications and advisory teams can inspect role alignment, skill evidence, ambiguity, and review guidance.
Classify role text against occupational intelligence, surface likely alignment, and expose weak or ambiguous input signals.
Extract and normalize skill signals from job text, then return structured evidence alongside quality and confidence context.
Use confidence, uncertainty, weak-signal detection, and review flags to keep consequential workforce decisions human-led.
Decision Passport
Add include_passport=true on supported grounded report calls to receive a compact package with decision scope, evidence condition, limitations, review requirements, and trace correlation.
Workforce Evidence Packs · Controlled beta
Select supported SOC occupations and NAICS industries for the approved U.S./Connecticut snapshot. Receive account-scoped JSON and standalone HTML with OEWS employment and wages, separate QWI industry flows, O*NET tasks and skills, source lineage, limitations, and employer-validation questions.
Each package keeps occupation, industry, and reference evidence separate, with explicit lineage and buyer-review boundaries.
AvelinLabs exposes workforce intelligence through authenticated JSON contracts. Submit role text, inspect ranked occupation results, and route uncertain cases for review.
curl -X POST "$BASE_URL/api/v1/job/analyze" \
-H "Authorization: Bearer $AVELIN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"title": "Business Analyst",
"description": "Facilitate stakeholder workshops, define requirements, and analyze data.",
"debug": false
}'
# Returns ranked occupations, skill signals,
# confidence, uncertainty, and review guidance.From role context and customer criteria to evidence-backed output for human review.
Start with job text and tenant-scoped role criteria or competency evidence available through Customer Grounding.
AvelinLabs combines occupational grounding, extracted skill signals, input-quality checks, and relevant customer evidence.
Structured results expose alignment, confidence, uncertainty, evidence references, and flags for further human review.
Technology and standards
Occupational grounding uses the O*NET® 30.3 Database. Integration surfaces include documented REST endpoints, an OpenAPI-described API surface, structured JSON responses, and an optional customer-operated MCP adapter.
Product and company names may be trademarks of their respective owners. Their use does not imply endorsement.
These decision-support lenses describe inspectable AvelinLabs outputs across role context, evidence quality, and human review.
Inspect likely occupation alignment, input quality, and ambiguity before treating a job description as a reliable role definition.
Structured role context
Trace grounded findings to evidence references and source context so reviewers can understand what influenced the output.
Support and provenance
Use confidence, uncertainty, weak-signal detection, and review flags to decide what needs further human judgment.
Guidance and escalation
A concise view of current scope and deliberate boundaries.
Start with the documented capability that can answer your workforce question.
Start from a job title and description.
Ground a role review in customer-provided criteria.
Use current structured REST contracts.
Workforce intelligence in practice
Explore practical examples of how AvelinLabs keeps occupational intelligence, labor-market evidence, customer context, and human review visible. Each case starts with a real decision question, preserves the boundaries between sources, and shows what must still be validated before a team acts.
Use the getting-started guide to move directly from setup to your first documented AvelinLabs workflow.
Self-service onboarding
Follow the documented setup path at your own pace.
Start with the setup guide, continue through the public documentation, and use runnable examples as you build.
