# Semantic talent search Intervieux lets employers search for candidates using a plain-language description instead of exact keyword filters, and switch between searching only your own applicant pool or the opt-in Intervieux network of candidates who consented to be discoverable. Keyword search on a resume database misses the candidate who described the same skill in different words, or whose strongest evidence lives in an interview transcript rather than a bullet point. Semantic search reads for meaning, not exact matches, so a description like someone who has led a technical team through a difficult migration can surface a candidate whose resume never uses those exact words. ## What it does Talent search takes a natural-language description of what you are looking for and matches it against candidate data using vector embeddings rather than keyword filters. A toggle switches the search scope between My talent pool, candidates already in your own pipelines, and the Intervieux network, candidates elsewhere on the platform who opted in and consented to being discoverable by employers based on their verified interview history. ## How it works 1. **Describe who you are looking for** — Write a plain-language description of the candidate profile you need, in the same language you would use describing the role to a colleague, not a string of exact-match keywords. 2. **Choose the search scope** — Toggle between My talent pool to search only candidates already in your own pipelines, or the Intervieux network to reach candidates elsewhere on the platform who consented to being discoverable. 3. **Semantic matching runs against the description** — The search compares your description against candidate data using vector embeddings, surfacing people whose actual profile matches the meaning of what you asked for, not just candidates whose resume happens to contain the same words. 4. **Network results respect candidate consent** — Any candidate surfaced from the network side has explicitly opted in and consented to being discoverable through a verified interview profile. Candidates who did not opt in do not appear in network search results at all. ## Meaning over keywords A candidate who wrote about debugging a production outage under pressure and a candidate who wrote about staying calm during a critical incident are describing something similar in different words. Semantic search built on vector embeddings can connect that description to both, where a keyword filter would only catch whichever phrase happened to match. ## Two pools, one search box My talent pool searches candidates already inside your own pipelines, useful for finding someone from a past role who might fit a new opening. The Intervieux network extends the same search out to candidates across the platform who chose to be found, expanding the pool beyond who has already applied to you directly. ## Consent is not an afterthought on the network side Network discoverability is opt-in. A candidate's verified interview profile only appears in employer search results if they explicitly consented to that, which keeps the network side of talent search grounded in candidates who actually want to be found, not a database scraped without permission. ## Results connect to the rest of the ATS A candidate surfaced through search is not a dead end. From My talent pool results, a recruiter can add someone to a curated candidate list or work them directly in the pipeline they already belong to. Network results give a starting point for outreach, since Intervieux's own invite links exist for exactly that purpose, getting a promising candidate into a real interview quickly. ## Who this is for This is for recruiters trying to fill a role faster than a fresh job posting allows, someone checking whether a past applicant fits a new opening, or a team wanting to reach beyond their own applicant history into a wider pool of candidates who have already completed verified AI interviews elsewhere on the platform. ## Frequently asked questions ### How is semantic search different from keyword search? Semantic search matches based on the meaning of your description using vector embeddings, so a candidate described in different words than your search can still surface, where a keyword filter would only catch an exact or near-exact match. ### What is the difference between My talent pool and the Intervieux network? My talent pool searches only candidates already inside your own pipelines. The Intervieux network extends the search to candidates elsewhere on the platform who opted in and consented to being discoverable through a verified interview profile. ### Can I see every candidate on the platform through network search? No. Only candidates who explicitly opted into network discoverability appear in those results. Consent is required before a candidate's profile is searchable by employers outside their own applications. ### Do I need to use exact job titles or skills to search? No. A plain-language description of the profile you need works, since the search matches on meaning rather than requiring precise keyword phrasing. ## Related pages - [Curated candidate lists](/features/curated-lists) - [Shareable interview invite links](/features/interview-invites) - [One-click candidate shortlisting](/features/shortlisting) - [Frequently asked questions](/faq) ## Describe who you need and let search find them Search your own talent pool or the opt-in Intervieux network with a plain-language description instead of exact keyword filters. Start practicing free: https://www.intervieux.ai/register · Hire with Intervieux: https://www.intervieux.ai/employers/signup