Skip to main content
A company page gives a native company ID. That ID can scope later operations. If you want a company’s newest jobs generally, use its jobs collection. If you want software-engineering roles at that company, use job search with both the title query and returned company ID; the first page of a broad jobs collection may contain no matching role.
The 1035 ID came from a Microsoft company read in a September 2026 study. In a new task, inspect the actual company result first; do not reuse a remembered ID for another organisation. That study’s broad company.jobs page held 25 of 2,152 reported jobs and no matching title; the filtered search returned three matching records. The broad read was a chosen detour, not a required step. The listings and prices, dates or availability of any job are a retrieval-time snapshot. These operations keep their native shapes inside the Almanac envelope. In that study, company jobs were at .data.data.items[], while filtered job search records were at .data.data[]. Check the current result’s type before scripting against either path. A null cursor on a resource-scoped page does not prove all 2,152 jobs were traversed. LinkedIn also has separate people, posts, articles, ads and location operations. A location filter may require a native geo ID rather than place text; inspect linkedin.location.search and the job-search contract for current fields. Search a person only when that is part of the task, and retain the returned original listing URL when reporting a job.