PhD, Postdoc & Research Jobs Scraper — 12 Sources in 1
Pricing
from $3.00 / 1,000 job results
Every PhD, postdoc and research job source behind one endpoint: EURAXESS, jobs.ac.uk, UN Careers, ReliefWeb, Impactpool, Devex, Ikerbasque, EURES, LinkedIn, AcademicPositions + 2 university boards. One run returns a merged, deduped dataset - billed once. Incremental mode surfaces only new postings.
Which boards to include. Leave empty to run the default set (all 10 boards that run inside this Actor — one flat bill, no extra fees). 'AcademicPositions' is the one exception: it needs a real browser to clear Cloudflare, so it runs as a separate child Actor and adds that Actor's own start + per-result fees on top of this bundle's. It is off by default; tick it only if you want it.
[ "euraxess", "jobs_ac_uk", "ikerbasque", "math_ku_phd", "ub_doctoral", "un_careers", "reliefweb", "impactpool", "linkedin", "eures"]Optional free-text filter forwarded to the boards that support it (the others ignore it). E.g. 'machine learning', 'climate', 'postdoc'.
Deprecated compatibility input. No active source requires an Anthropic key for full-description extraction.
Deprecated compatibility input. No active source requires a Mistral key for full-description extraction.
Deprecated compatibility input; currently unused.
Deprecated compatibility input. No active source requires an OpenAI key for full-description extraction.
Deprecated compatibility input; currently unused.
Cap on items fetched from EACH board before merging. Raise it for a deeper sweep of every board; the merged total is still bounded by 'Max items (total)' below.
Hard cap on the merged, deduped output — and therefore on what you pay, since this Actor charges per unique job returned. Default (240) is sources × 'Max items per source'. Set 0 to remove the cap; cost then scales with every board's raw output.
How long to reuse results already fetched from a board instead of fetching again. Set 0 to always fetch fresh data.
How many boards to fetch at the same time. The default fetches them all at once, so a run takes as long as the slowest board rather than the sum.
How long to wait for each board before giving up on it and merging what the others returned. Boards run in parallel, so this is a per-board ceiling, not a budget shared across them.
Leave empty — injected automatically when this Actor runs on the Apify platform. Only set this when running the code outside the platform, e.g. on your own machine.