Skip to main content

Rate limits

Limits apply to each API key, not to your workspace or your IP address.

LimitValue
Requests per key60 per minute
Extract candidate details10 per minute, because each file is read by a model
Get a job's counts30 per minute, because it reads a job's whole application history
Get an interview's report30 per minute, the same limit as a job's counts
CV files per extract request10
Candidates per submit request50

These are the starting values for v1 and may change before launch. The response headers are always the source of truth for your key.

Headers​

Successful responses report where you stand:

HeaderMeaning
X-RateLimit-LimitRequests allowed in the current window.
X-RateLimit-RemainingRequests left in the current window.
X-RateLimit-ResetSeconds until the window resets.

When you go over, the response is 429 rate_limited with a Retry-After header, in seconds.

Backing off​

Wait for Retry-After, then retry. This helper does that and gives up after three attempts. It is safe to use with Submit candidates as long as you reuse the same Idempotency-Key.

# curl retries 429 and 5xx on its own, and honours Retry-After.
curl --retry 3 -X POST https://api.tarkflo.com/v1/candidates/submit \
-H "Authorization: Bearer $TARKFLO_API_KEY" \
-H "Idempotency-Key: 6f0a2d4e-9c1b-4f7a-8e35-1b2d3c4e5f60" \
-H "Content-Type: application/json" \
-d @batch.json