Shortlist jobs from posting text after counting the market
At a glance
Kevin Hall connected Claude Code to Apify and tested a search that selected eight backend roles using the actual posting text. This workflow is listed under Job Search, Research.
- Agents / tools
- Claude Code, Apify MCP
SourceBuilding a tool for these workflows? See weekly rates.
What they did
1) Have Claude Code and an Apify account. Add the MCP connection with `claude mcp add --transport http apify "https://mcp.apify.com?tools=b2bsearch/linkedin-job-search-api,b2bsearch/hiring-companies-finder,b2bsearch/linkedin-jobs-scraper"` and authenticate. 2) Provide your experience, stack, country, seniority, and employer-size range. Request a count first using `previewOnly: true`; narrow the search if it exceeds 400. 3) Request 40 recent postings with full text, agency filtering, and hiring pace. Retrieve rows with `get-dataset-items` using the returned dataset ID. Keep the newest row per company ID and title. 4) Ask Claude to read the texts and return eight matches with supporting stack quotations. Review those quotations and experience requirements before choosing applications.
How it went
The reported session took three minutes and 11 turns; six shortlisted roles had fewer than 25 applicants.
Worth knowing
On the test date, the database lagged six days. The Actors did not support an exact remote-work filter.
