Structured search intent

The intent object describes who you want to find when your application already has structured criteria. Send it alone for a fully structured request, or alongside top-level query when plain English provides useful context and specific fields need to be corrected or constrained. When both are present, Orbit derives a base intent from query and overlays the caller-provided intent. At each supplied field, arrays, scalar values, and null replace the derived value; nested objects merge field by field.

{
  "intent": {
    "experiences": [
      {
        "titleAnyOf": ["Machine Learning Engineer", "ML Engineer"],
        "organization": "OpenAI",
        "temporalScope": "current"
      }
    ],
    "semanticClauses": [
      { "text": "works on large language model inference" }
    ],
    "geo": {
      "place": "San Francisco Bay Area",
      "distance": "50 miles"
    }
  }
}

Do not send intent.version. Orbit owns and applies the schema version internally.

Top-level fields

Field	Type	Description
`names`	string[]	One or more person names.
`experiences`	object[]	Job title, organization, dates, and current/historical constraints.
`semanticClauses`	object[]	Free-form professional criteria that do not fit an exact field.
`schools`	object[]	School, dates, graduation year, or relationship.
`geo`	object or `null`	Place, distance, and historical-location constraint.
`demographics`	object or `null`	Age, birth year, or gender constraints.
`personalization`	object or `null`	First-degree network or near-me constraints.

Unknown and server-owned fields are rejected.

Experience

{
  "experiences": [
    {
      "title": "Founder",
      "titleAnyOf": ["Founder", "Co-founder"],
      "organization": "Example Co",
      "year": 2024,
      "startYear": 2020,
      "startYearLte": 2021,
      "endYear": 2025,
      "temporalScope": "current"
    }
  ]
}

temporalScope accepts current, historical, or both. Use either title or titleAnyOf when possible; sending both is allowed but usually redundant.

Semantic clauses

{
  "semanticClauses": [
    { "text": "built production recommendation systems" },
    { "anyOf": ["robotics", "autonomous vehicles"] }
  ]
}

Use text for one concept or anyOf for alternatives.

Schools

{
  "schools": [
    {
      "schoolAnyOf": ["Stanford University", "UC Berkeley"],
      "graduationYear": 2020,
      "temporalScope": "historical",
      "relation": "attended"
    }
  ]
}

School objects accept school, schoolAnyOf, graduationYear, startYear, endYear, temporalScope, and relation.

Location and demographics

{
  "geo": {
    "place": "New York City",
    "isHistorical": false,
    "distance": "25 miles"
  },
  "demographics": {
    "ageRange": { "min": 30, "max": 45 },
    "birthYearRange": { "min": 1980, "max": 1995 },
    "gender": "woman"
  }
}

Each range requires both min and max, and min must not exceed max.

Personalization

{
  "personalization": {
    "network": { "scope": "first_degree" },
    "nearMe": { "distance": "10 miles" }
  }
}

See Start a search for the complete request contract.

Search

Enrich

Watchers

Webhooks

Keys

Structured search intent

Top-level fields

Experience

Semantic clauses

Schools

Location and demographics

Personalization

​Top-level fields

​Experience

​Semantic clauses

​Schools

​Location and demographics

​Personalization

Top-level fields

Experience

Semantic clauses

Schools

Location and demographics

Personalization