> ## Documentation Index > Fetch the complete documentation index at: https://docs.datamagnet.co/llms.txt > Use this file to discover all available pages before exploring further. # Changelog: September 2026 > September 2026 Datamagnet API release notes: the Person response now has the same fields on every request, several fields change type or name, and many new profile fields are added. ## Week of September 19 ### Person: the same response shape on every request [Person](/api-reference/endpoints/people) now returns the same set of fields, in the same order, on every request. Every entry in a list (`experience`, `education`, `certification`, `publication`, `project`, `volunteering`, `honors_and_awards`, `organizations`, `patents`, `language`, `courses`) has the same keys. Before, a field could be present on one request and missing on the next for the same profile. That no longer happens, including for profiles served from cache. A field with no value is still present: * `""` for text * `[]` for lists * `null` where the value is not known Check a field's value, not whether the key exists. ### Breaking changes | Field | Before | Now | | ---------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------ | | `followers`, `connections` | sometimes `0` when the count was not available | `null` when the count is not available | | `job_still_working` (profile and `experience[]`) | `false` for a role with no dates | `null` for a role with no start or end date; `true` / `false` only when the dates are known | | `certification[].startedOn` | `startedOn` | renamed to `started_on` (same object) | | `publication[].published_on` | sometimes an object `{ "year", "month", "day" }` | always the string `"M/D/YYYY"` (`""` when there is no month); the object is now in `published_on_date` | | `honors_and_awards[]`, `organizations[]`, `patents[].inventors[]`, `project[].contributors[]`, `courses[]` | sometimes camelCase keys (`issuedOn`, `positionHeld`, `firstName`, …) and `multiLocale*` copies | snake\_case keys only (`issued_on`, `position_held`, `first_name`, …); `multiLocale*` removed | | `location_state`, `location_county` | sometimes `"City, State"` | `location_state` is the state or region; `location_county` is always `""` | | `experience[].company_id` | `"0"` for a company without a LinkedIn page | `""` | | `experience[].job_location_country` | could end in `" · Remote"` | the country only; the workplace type is in the new `workplace_type` | | `honors_and_awards[].issuer` | sometimes `"Issued by …"` | the issuer's name only | | `project[].description` | could start with `"Associated with …"` | that line is in `associated_with` | | `volunteering[].is_current_position` | `true` for an entry with no dates | `true` only when the entry has a start date and no end date | ### New fields **Profile** * **Location:** `address_without_country`, `address_country_only`, `country_code` (ISO 3166 alpha-2, e.g. `"US"`) and `location_city`. The existing `location`, `country`, `address_with_country` and `address` are now present on every response too. * **Industry:** `industry` and `industry_name` (always the same value), plus `industry_urn`. * **Pictures:** `profile_pictures` / `avatar_all_dimensions` and `background_pictures` / `background_picture_all_dimensions`. Both forms are always present and list the same images. * **Other fields now present on every response:** `total_experiences`, `primary_locale`, `is_hiring`, `is_memorialized`, `show_premium_badge`, `courses`, `updates`, `contact_info`. A flag that could not be read is `null`, not `false`. * **Birthday:** `birthday` is filled with the month and day when the profile shows them. **Experience** * `workplace_type`: `"On-site"`, `"Remote"` or `"Hybrid"`, when the role shows it. * `company_username` and `company_urn` on every role. * When a company is known to us, `company_id`, `company_username` and `company_url` identify it the same way on every request. `company_url` takes the form `https://www.linkedin.com/company//`. * `company_headcount_range` always uses the form `"11-50 employees"`. **Other sections** * `education[].university_urn`. * `publication[]`: `published_on_date` (`{ "day", "month", "year" }`, `0` when unknown), `publishing_date` (`"Nov 11 2014"`) and `authors`. * `project[]`: `start`, `end`, `date`, `url`, `contributors`, `associated_with` and `subtitle`. * `volunteering[]`: `started_on`, `ended_on`, `start_date`, `end_date`, `duration`, `is_current_position`, `description`, `cause`, and `organization` alongside `company_name`. * `certification[]`: both naming styles are always filled, `issuer` / `authority` / `company_name`, `issued_on` / `started_on` and `expires_on` / `end_date`. * `organizations[]`: `name` / `title`, `position_held` / `subtitle`, `start` and `end`. * `honors_and_awards[].issued_on`. ### Fixes * Education dates written with months (e.g. `"Sep 2023 - Present"`) were being dropped. They are now returned, including the month. * Experience lists for profiles with a long work history are more complete. * `use_cache: "never"` now always fetches the profile fresh. Before, repeating the same request within a few minutes could return the previous response. * A role whose only location was its workplace type (e.g. "Remote") now gets the profile's location, like any other role without one.