This video is available to ERE Pro subscribers

Subscribe to watch

Already a member? Log in

GAI, Boolean, and Natural Language Search- A Collision Course for The Future

October 22, 2024 · SourceCon Fall 2024 ·

Speakers

About this video

Boolean search strings get run through a lineup of free generative AI tools to see which ones can actually replicate what a LinkedIn Recruiter seat does, and which ones just hand back generic search advice. The results are inconsistent in useful ways: some tools refuse to play along, others quietly deliver profiles, summaries, and even sourcing beyond LinkedIn's paywall.

The core method starts with a classic X-ray search, site:linkedin.com/in "account manager" "Verizon" OR "Salesforce", then drops that same string into ChatGPT, Claude, Meta AI, Gemini, Perplexity, and you.com to compare what comes back. A few produce nothing but instructions on how to search LinkedIn manually. Others return actual candidate photos, summaries of responsibilities, and footnoted links straight to profiles.

Key tactics covered include:

  • Layering natural language phrases like "responsible for" or "exceeded quota by" into Boolean strings to surface overachievers and leadership language on resumes
  • Using job descriptions from a competitor's site to figure out what to ask a candidate currently working there
  • Asking AI tools to identify soft skills, synonyms, competitor company lists, acronyms, and common misspellings tied to a target job title
  • Comparing outputs side by side using a search engine that lets multiple paid AI models run the same query for a flat low cost
  • Uploading resumes, job descriptions, and even video or audio content into a notebook-style AI tool that summarizes documents, ranks candidates against each other, and generates a spoken discussion of who fits a role best
  • Using that same tool to prep for hiring manager intake meetings by walking in with pros and cons already mapped out for a shortlist of candidates

The throughline is curiosity paired with tenacity: testing multiple tools against the same query, refining prompts when a model pushes back, and treating free AI access as a legitimate way to reach candidates who would otherwise sit behind a subscription paywall.

GAI, Boolean, and Natural Language Search- A Collision Course for The Future | ERE Pro