HR & recruiting

LinkedIn's Hiring Assistant cuts recruiters' review time by 48%

By LinkedIn (David Tag, Prashanthi Padmanabhan)
LangGraphLangChainPython

Built and shared by LinkedIn (David Tag, Prashanthi Padmanabhan), not by me. Summarised from the public sources below. Check them before you rely on a number.

Who
LinkedIn (David Tag, Prashanthi Padmanabhan)
Category
HR & recruiting
Source
YouTube · Web
Dates
Published 13 Jun 2025 · Found 17 Oct 2026

LinkedIn built Hiring Assistant, its first AI agent in production, which takes the heavy lifting of finding and evaluating candidates off recruiters. David Tag from LinkedIn's AI team explains the architecture, and LinkedIn reports that users spend about half as much time reviewing applications.

What they built

A supervisor agent coordinates sub-agents that call LinkedIn's services through 'skills' in a shared skill registry. LinkedIn standardised its generative AI work on Python and LangGraph and built a shared service framework and agent platform with asynchronous, event-driven messaging, now used by more than 20 teams. The agent also connects to companies' applicant tracking systems, so it can evaluate the whole candidate pool and not just LinkedIn profiles.

The result

Users spend 48% less time reviewing applications, review 62% fewer profiles before deciding on a candidate, and get a 69% higher InMail acceptance rate than with traditional sourcing (self-reported by LinkedIn).

Sources

Next step

Want this in
your business?

Tell me which case caught your eye. I will tell you straight whether it fits, and what it would take.

Start a conversation →