API reference / Devices, signing & primitives / Warm device feed

Warm device feed

Calibrate a provisioned guest device context toward a niche with organic feed pulls and AppLog telemetry.

POSTapi.tikultra.com/v1/devices/{id}/warmHeavy · 10 credits

Upstream: /aweme/v1/feed//service/2/app_log/
Access: Growth and up
Runs on: Primitive
Billing: Failed calls are free

Authorization & parameters

x-api-keyheader · stringrequired

Your API key. Create one in API keys.

Path parameters

idstringrequired

The device ID. For example 7392011847562901234.

Body

Send JSON with content-type: application/json.

nichestringrequired

Topic or keyword cluster to calibrate toward. For example mechanical keyboards.

duration_secinteger

Warming window in seconds. For example 45.

Client and egress

Runs locally on the protocol primitive layer with deterministic codec / cryptographic state. See Devices, signing & primitives.

Response

A 200 returns JSON with stable field names. Every response carries x-credits-cost, x-credits-tier and x-credits-remaining.

Response fields

Key top-level and nested fields decoded from the mobile upstream payload:

statusstring

Current lifecycle or health status.

context_idstring

Isolated device/proxy context identifier in LeanPool or FullPool.

device_idstring

Provisioned mobile device identifier.

regionstring

Two-letter ISO 3166-1 country code associated with the item or context.

citystring

City-level egress or location identifier.

nichestring

Decoded niche value.

feed_pullsinteger

Decoded feed pulls value.

videos_dwelledinteger

Decoded videos dwelled value.

applog_batches_flushedinteger

Decoded applog batches flushed value.

warm_tierstring

Interest-graph calibration state (cold, warming, calibrated_warm).

took_msinteger

Gateway execution or signing latency in milliseconds.

Errors

Errors return a JSON body with a stable code and cost 0 credits. See Errors and retries.

import os from tikultra import TikUltra with TikUltra(api_key=os.environ["TIKULTRA_KEY"]) as client: res = client.request( "POST", "/v1/devices/7392011847562901234/warm", json_body={ "niche": "mechanical keyboards", "duration_sec": 45, }, ) print(res.credits_cost, res.data)

{ "status": "warmed", "context_id": "ctx_9mK4vP2x", "device_id": "7392011847562901234", "region": "US", "city": "miami", "niche": "mechanical keyboards", "feed_pulls": 3, "videos_dwelled": 8, "applog_batches_flushed": 2, "warm_tier": "calibrated_warm", "took_ms": 42810 }