Post to LinkedIn, call in-app, and ask your ATS from Claude

01Post your jobs to LinkedIn#

Posting a role used to mean rebuilding it inside LinkedIn by hand: copy the description across, keep two diverging versions in sync, and watch applicants pile up in a LinkedIn inbox instead of your pipeline. LinkedIn job posting removes the double entry.

The listing points back at your Hivemind application form, so every applicant clicks through and enters the pipeline you already built, with your screening, assessments and automations applied from the first minute. One listing, one workflow, nothing entered twice.

1. 1Connect your LinkedIn account under Settings → Integrations.
2. 2Open a job and choose Post to LinkedIn.
3. 3Applicants click through to your application form and enter your pipeline automatically.

02Phone screens that file themselves#

Phone screens used to happen off-platform: dial from a personal phone, take notes in a doc, and leave nothing on the candidate record. The screen happened, but the system never knew. Standalone calling puts the screen back on the record.

Open a candidate and hit Call. There is no phone system to set up. The call is recorded and transcribed automatically, both attach to the candidate record the moment you hang up, and past calls are searchable by candidate name.

Calls are billed from your credit balance by duration, with no separate telephony contract or per-seat line; current rates are on the pricing page. Hiring managers benefit too: instead of a second-hand summary, they read the transcript of the actual conversation.

03Your ATS, inside your AI assistant#

The Hivemind MCP server connects your hiring data to the AI assistant your team already uses: Claude, ChatGPT, Microsoft Copilot, or any client that speaks the Model Context Protocol, over stdio or streamable HTTP.

It registers 25 tools covering pipelines, candidates and assessment results, secured by your Hivemind API key and scoped to your company. Ask "who is still in screening for the backend role?" or "how is the senior pipeline converting?" in plain language and the answer comes grounded in your live data.

Sessions are read-only: your assistant can look anything up, and it can never change anything in your account.

1. 1Generate an API key in Hivemind under Settings → Integrations.
2. 2Add the Hivemind MCP server to your AI assistant, over stdio or streamable HTTP.
3. 3Ask about your pipelines, candidates and assessment results in plain language.

04Resume fact screening#

The first pass over a stack of applicants is usually a person checking the same few facts on every resume: the right degree, the years of experience, the must-have skill. Resume fact screening turns those checks into pipeline logic.

Define true or false facts to screen on, and candidates are routed down different pipeline branches automatically based on the answers. The ones who clear the bar move straight to the next stage, the rest take whichever path you designed for them, and nobody spends an afternoon on the first pass.

05Smarter prospect search#

Job titles are written a hundred different ways, and an exact-match search misses most of them. Prospect search now matches related job titles too, so a search for "backend engineer" also surfaces the people who wrote it differently.

You also get a look at the search criteria before anything runs. Review what the search is about to match on, adjust it where it misses your intent, and then run it.

06Company pages go live#

Company pages built in the page builder can now be published. Assemble the sections of your careers page, put them in the order you want, pick your brand color, and take the page from draft to live without leaving Hivemind.

The published page gets its own address on your company slug, ready to be the destination for your job listings and outreach.