New MCP server connects AI assistants to verified candidate data, with shortlists sent straight to Bullhorn and Airtable
LONDON and ARLINGTON, Texas | November 9, 2026 | PitchMe, the data intelligence layer for recruitment and staffing agencies, today launched a remote Model Context Protocol (MCP) server that lets recruiters run candidate searches through ChatGPT, Groq and Muse. Recruiters can describe a role in plain language and ask the assistant to search PitchMe, build a shortlist, and reveal available contact details, all within their own PitchMe account.
Recruiters already use general-purpose AI assistants to draft job ads, conduct outreach and create meeting notes. PitchMe’s MCP server connects those assistants to its candidate data and sourcing tools, allowing recruiters to move from asking questions to taking action.
For example, a recruiter can ask: “Find ICU nurses in Texas with active state licenses, shortlist the top 20 and reveal their contacts.” The assistant runs the search in PitchMe and returns a shortlist the recruiter can act on.
“Recruiters don’t need another tab. They need the tools they already use to do more of the work,” said Dina Bay, founder and CEO of PitchMe. “But an AI assistant is only as useful as the data underneath it. By connecting AI assistants to verified candidate data, we’re turning a prompt into a shortlist that recruiters can actually call, already filed in their ATS, with the agency in control of who can see what.”
Built for agency control. The assistant works only within the account, team, and permissions the recruiter already holds in PitchMe. Users choose how long a connection lasts, including time-limited sessions or access until revoked, and can disconnect it at any time. PitchMe is SOC 2 certified and GDPR compliant.
Healthcare provider data. For healthcare staffing agencies, the assistant can also search the US national provider dataset (NPPES). Matched candidate records show NPI number, credentials, specialty, and license states.
“The agent performs actions on the recruiter’s behalf, and that work is visible inside PitchMe,” said Max Kuzkin, PitchMe’s CTO. “Recruiters can inspect the search criteria, refine their instructions and see the project and shortlist as they develop.”
From search to ATS in fewer steps. The MCP server launches alongside updates that shorten the path from finding a candidate to placing them:
Shortlist & Reveal: recruiters save an external candidate from search results and request their available contact details in one step, or run reveals across a whole project shortlist.
Bullhorn and Airtable write-back: Once the agency’s integration and field mapping are configured, recruiters can instruct the assistant to send revealed candidates to an open Bullhorn job or Airtable table. Candidates already on that job or table are excluded automatically.
Additional improvements include CSV and Excel candidate imports, self-service team joining through verified company domains, and expanded search coverage in Belgium, Bulgaria, Luxembourg and the Netherlands.
Today’s release follows PitchMe’s launch of SONAR, its unified search across an agency’s own database and the external talent market, at Staffing Industry Analysts’ CollaborationX conference in Dallas in September.
Availability. The MCP server is live now for PitchMe customers and works with ChatGPT, Groq, and Muse. Setup requirements vary by assistant. Agencies can request a demonstration at pitchme.co. Setup guides are available at: https://pitchme.co/docs/sdk/use-the-mcp-server
About PitchMe
PitchMe is the data intelligence layer for recruitment and staffing agencies. It enriches and verifies candidate data inside the systems agencies already run, including Bullhorn and Avionte, and its SONAR search engine brings an agency’s own talent pool and the external market into a single search. PitchMe serves agencies across the UK, US, and Europe, with a growing focus on healthcare staffing. Learn more at pitchme.co.