| capability | LinkFetch | Bright Data |
|---|---|---|
01Jobs floor price | $0 · $5 credit, then $29 / mo | $250 min order |
02Per-call cost (jobs) | 1 credit per page, empty = free | $2.50 / 1K records |
03Bulk export | $199 one-time · 8M+ rows | $250 min order |
04Typed jobs schema (parsed salary, level, applicant count) | yes | Range only |
05is_active flag re-checked on every crawl | yes | no |
06Applicant count + timestamped history | yes | no |
07Real-time webhook | Not yet | Yes |
08MCP-native (Claude · Cursor · Zed) | yes | no |
09Time to first call | 90 sec | Hours · proxy setup |
10User-is-principal model (profiles) | yes | no |
11No proxy bandwidth fees | yes | no |
12Provenance on every row | yes | no |
13GDPR + CCPA aligned | yes | Partial |
14Delivery formats | Postgres dump · CSV · JSONL · live DB | S3 · GCS · Snowflake · Pub-Sub |
15Multi-source dedupe (LinkedIn + Indeed + Glassdoor) | no | yes |
LinkFetch: typed API, provenance on every row
Bright Data: proxy + parser workflow you maintain
# Bright Data — LinkedIn Jobs Dataset
# $250 minimum, snapshot triggered, fetched on completion
const trigger = await fetch("https://api.brightdata.com/datasets/v3/trigger", {
method: "POST",
headers: { Authorization: `Bearer ${BD_KEY}` },
body: JSON.stringify({
dataset_id: "linkedin_jobs_dataset",
filter: { country: "US", role: "backend engineer" },
}),
});
// poll the snapshot id, download CSV, parse, dedupe yourself// LinkFetch — typed jobs API, no minimum, plain fetch
const params = new URLSearchParams({
q: "backend engineer",
location: "United States",
posted_within: "week",
});
const res = await fetch(`https://api.linkfetch.io/v1/jobs?${params}`, {
headers: { Authorization: `Bearer ${process.env.LINKFETCH_KEY}` },
});
const { data } = await res.json();
// rows already typed, deduped, with is_active + applicant_countYou shouldn't, if you're an indie dev or small team. Bright Data's LinkedIn Jobs Dataset is sold as bulk: $250 minimum order, $2.50 per 1,000 records. LinkFetch is per-call: 1 credit per search (~$0.005 at the Starter tier), no minimum, no snapshot polling. For under ~100K records/mo, LinkFetch is cheaper and the rows are already typed.
Not yet. Bright Data can push a finished snapshot to your endpoint, but you still pay per record and receive the raw scraped row. With LinkFetch you poll GET /v1/jobs with posted_within=24h (1 credit per page, typed rows), or take the $49/month live database and query first_seen_at directly. Push delivery is on our roadmap and will be announced in the changelog when it ships.
If you're feeding jobs into an agent's index, you'll fill it up with expired listings within a month. We re-check is_active on every recrawl and surface it as a boolean, so your index stays clean. Bright Data exposes job_posted_date and dataset refresh metadata, but you have to derive expiry yourself.
Bright Data sells proxies; customers are responsible for what they fetch and how. For LinkedIn profiles specifically, they ship pre-built scrapers that fetch behind logins using rented accounts — which violates LinkedIn's terms and has led to repeated enforcement actions. LinkFetch doesn't have this problem because there are no rented accounts in our model — profile reads run on the user's own session via the Chrome extension, jobs come from public guest endpoints.
Bright Data is a general-purpose web-scraping platform; LinkFetch is LinkedIn-specialised. If you need Amazon / Walmart / Google SERP / whatever, stay with Bright Data (or another vertical specialist). If your workload is LinkedIn-centric — jobs, profiles, companies — LinkFetch is cleaner and cheaper.
Yes. Common pattern: LinkFetch for LinkedIn jobs, profiles, and companies via the typed API; Bright Data for everything else (Amazon, Google SERP, generic web scraping). They don't conflict.