Machine-Observed Work Histories Replace Resumes

The core bet is that professional identity will move from self reported resumes to machine observed work history. If AI tools sit inside email, meetings, project trackers, and hiring systems, they can build a live record of what someone actually shipped, who they worked with, and how they performed. That weakens the value of a profile page built on manual updates and keyword optimization.

• LinkedIn is still huge and getting more automated, which shows how large this opportunity is. Microsoft said LinkedIn passed $15B in revenue and 950M members in fiscal 2023, and LinkedIn now sells Hiring Assistant as an add on to Recruiter that sources, evaluates, and ranks candidates for employers.
  2 microsoft 3 linkedin 4 linkedin
• The newer challengers are already built around inferred skills instead of a static profile. Findem uses 3D profiles for external and internal talent, Juicebox searches 800M profiles across many sources, and Mercor monetizes matching through a 30% placement fee, turning talent data into workflow and transaction revenue.
  5 sacra 6 sacra 7 sacra
• What changes is not just candidate search, but the proof layer underneath it. The interview points to AI as middle management, meaning the same system assigning work and following up can also log outcomes. That creates a richer record than endorsements, job titles, or cold outbound messages on a network site.
  1 sacra

The next step is a shift from networking products to talent operating systems. The winning platforms will own the system of record for what work was done, then use that data to fill jobs, staff projects, benchmark skills, and route opportunities automatically. In that world, the profile becomes a byproduct, not the product.