> ## Documentation Index
> Fetch the complete documentation index at: https://nova.dweet.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Get Scoring Latency
> Time-series of P50, P95, and P99 scoring duration percentiles
Returns percentile breakdowns of end-to-end scoring duration over time. Useful for tracking performance trends and identifying latency spikes.
This endpoint requires only a valid API key. `X-Tenant-Id` isn't required.
## Query parameters
| Parameter | Required | Values | Description |
| ----------- | -------- | ------------------------------- | ---------------------------- |
| `timeRange` | Yes | `1h`, `24h`, `7d`, `30d`, `90d` | Lookback window for the data |
## Example response
```json theme={null}
{
"period": {
"start": "2026-03-07T00:00:00.000Z",
"end": "2026-03-08T00:00:00.000Z"
},
"granularity": "hour",
"data": [
{
"timestamp": "2026-03-07T12:00:00.000Z",
"p50Ms": 800,
"p95Ms": 2000,
"p99Ms": 3500
},
{
"timestamp": "2026-03-07T13:00:00.000Z",
"p50Ms": 750,
"p95Ms": 1800,
"p99Ms": 3200
}
]
}
```
## Response fields
| Field | Type | Description |
| ------------------ | -------------- | ------------------------------------------------ |
| `period.start` | string | ISO 8601 start of the period |
| `period.end` | string | ISO 8601 end of the period |
| `granularity` | string | Bucket size for each data point |
| `data[].timestamp` | string | ISO 8601 timestamp for this bucket |
| `data[].p50Ms` | number \| null | Median scoring duration in milliseconds |
| `data[].p95Ms` | number \| null | 95th percentile scoring duration in milliseconds |
| `data[].p99Ms` | number \| null | 99th percentile scoring duration in milliseconds |
P50 reflects the typical scoring experience. P95 and P99 surface tail latency that affects a small fraction of requests but can indicate bottlenecks.
Responses are cached for 60 seconds (`Cache-Control: private, max-age=60`). This endpoint is rate-limited under the `analytics` bucket (20 req/s).
## OpenAPI
````yaml GET /v1/analytics/scoring-latency
openapi: 3.1.0
info:
title: Nova Embed API
description: >-
The Nova Embed API enables ATS platforms to generate job-scoped screening
criteria and score applications asynchronously against those criteria.
version: 1.0.0
contact:
name: Nova Support
email: nova@dweet.com
servers:
- url: https://embed.nova.dweet.com
description: >-
Environment is determined by API key prefix: sk_test_* for sandbox,
sk_live_* for production.
security:
- bearerAuth: []
tags:
- name: Analytics
description: View scoring performance, webhook health, and error breakdowns.
- name: Criteria
description: Generate and manage job-scoped screening criteria.
- name: Scoring
description: Submit applications for scoring and fetch scoring job results.
- name: Deletion
description: Request application or tenant deletion and poll for completion.
- name: Criteria Library
description: Manage tenant-scoped reusable criteria templates.
- name: Rate Limits
description: Check your current rate limit status.
- name: Webhooks
description: >-
Events sent to your registered webhook endpoints. Verify signatures with
HMAC-SHA256 using your webhook secret.
paths:
/v1/analytics/scoring-latency:
get:
tags:
- Analytics
summary: Get scoring latency percentiles
description: >-
Returns time-series P50/P95/P99 scoring duration percentiles. Does not
require X-Tenant-Id.
operationId: getAnalyticsScoringLatency
parameters:
- name: timeRange
in: query
required: true
schema:
$ref: '#/components/schemas/TimeRange'
description: Time range for the analytics query
responses:
'200':
description: Scoring latency percentiles
headers:
X-RateLimit-Bucket:
$ref: '#/components/headers/X-RateLimit-Bucket'
X-RateLimit-Limit:
$ref: '#/components/headers/X-RateLimit-Limit'
X-RateLimit-Remaining:
$ref: '#/components/headers/X-RateLimit-Remaining'
X-RateLimit-Reset:
$ref: '#/components/headers/X-RateLimit-Reset'
content:
application/json:
schema:
$ref: '#/components/schemas/AnalyticsScoringLatencyResponse'
'400':
$ref: '#/components/responses/ErrorResponse'
'401':
$ref: '#/components/responses/ErrorResponse'
'429':
$ref: '#/components/responses/RateLimitedResponse'
'500':
$ref: '#/components/responses/ErrorResponse'
x-codeSamples:
- lang: typescript
label: '@nova-sdk/api'
source: >-
import { Nova } from "@nova-sdk/api";
const nova = new Nova({
apiKey: "sk_test_...",
tenantId: "acme-corp",
});
const latency = await nova.analytics.scoringLatency({ timeRange:
"24h" });
- lang: bash
label: cURL
source: >-
curl -X GET
"https://embed.nova.dweet.com/v1/analytics/scoring-latency?timeRange=24h"
\
-H "Authorization: Bearer sk_test_..."
components:
schemas:
TimeRange:
type: string
enum:
- 1h
- 24h
- 7d
- 30d
- 90d
description: Time range for analytics queries.
AnalyticsScoringLatencyResponse:
type: object
required:
- period
- granularity
- data
properties:
period:
$ref: '#/components/schemas/AnalyticsPeriod'
granularity:
type: string
enum:
- minute
- hour
- day
description: Time bucket granularity
data:
type: array
items:
$ref: '#/components/schemas/ScoringLatencyDataPoint'
AnalyticsPeriod:
type: object
required:
- start
- end
properties:
start:
type: string
format: date-time
description: Period start (ISO 8601)
end:
type: string
format: date-time
description: Period end (ISO 8601)
ScoringLatencyDataPoint:
type: object
required:
- timestamp
- p50Ms
- p95Ms
- p99Ms
properties:
timestamp:
type: string
format: date-time
description: Bucket timestamp (ISO 8601)
p50Ms:
type:
- number
- 'null'
description: 50th percentile scoring duration in ms
p95Ms:
type:
- number
- 'null'
description: 95th percentile scoring duration in ms
p99Ms:
type:
- number
- 'null'
description: 99th percentile scoring duration in ms
HttpError:
type: object
required:
- type
- code
- status
- message
- retryable
- traceId
properties:
type:
type: string
description: URI reference that identifies the error type
code:
$ref: '#/components/schemas/ErrorCode'
status:
type: integer
description: HTTP status code
message:
type: string
description: Error message
retryable:
type: boolean
description: When true, retrying the request may succeed
traceId:
type: string
description: Trace ID for debugging
details:
type:
- array
- 'null'
description: Field-level validation details
items:
type: object
required:
- field
- code
- message
properties:
field:
type: string
code:
type: string
message:
type: string
ErrorCode:
type: string
description: >-
Machine-readable error code. Generated from the canonical error
registry.
enum:
- UNAUTHORIZED
- FORBIDDEN
- VALIDATION_ERROR
- ANSWER_MISMATCH
- CRITERIA_INVALID
- CRITERIA_REVISION_CONFLICT
- NOT_FOUND
- TENANT_NOT_FOUND
- JOB_NOT_FOUND
- APPLICATION_NOT_FOUND
- QUESTION_SET_NOT_FOUND
- CRITERIA_NOT_FOUND
- CRITERION_NOT_FOUND
- CRITERIA_VERSION_NOT_FOUND
- SCORING_JOB_NOT_FOUND
- BATCH_NOT_FOUND
- LIBRARY_CRITERION_NOT_FOUND
- RATE_LIMITED
- MONTHLY_TRIAL_QUOTA_EXCEEDED
- IDEMPOTENCY_KEY_ALREADY_USED
- IDEMPOTENCY_REQUEST_IN_PROGRESS
- RESUME_FETCH_FAILED
- RESUME_CONVERSION_FAILED
- RESUME_ENCRYPTED
- RESUME_CORRUPTED
- RESUME_PARSE_FAILED
- RESUME_TOO_LARGE
- RESUME_EMPTY
- RESUME_URL_BLOCKED
- AI_PROCESSING_FAILED
- AI_GENERATION_FAILED
- AI_SCORING_FAILED
- MISSING_JOB_DESCRIPTION
- TIMEOUT
- MAX_RETRIES_EXCEEDED
- TASK_STUCK
- DELETION_REQUEST_NOT_FOUND
- DELETION_REQUEST_FAILED
- INTERNAL_ERROR
- SERVICE_UNAVAILABLE
- UNKNOWN
headers:
X-RateLimit-Bucket:
description: Which rate limit bucket the request was classified into
schema:
type: string
enum:
- criteria_ai
- criteria_current_reads
- scoring_intake_batch
- scoring_intake_single
- read_and_ops
- rate_limit_status
- analytics
X-RateLimit-Limit:
description: Maximum requests per second for this bucket
schema:
type: integer
X-RateLimit-Remaining:
description: Requests remaining in the current 1-second window
schema:
type: integer
X-RateLimit-Reset:
description: Unix timestamp when the current window resets
schema:
type: integer
Retry-After:
description: >-
Seconds to wait before retrying a short-window RATE_LIMITED or
service-unavailable response whose body has retryable: true.
schema:
type: integer
X-RateLimit-Degraded:
description: >-
Present and set to "true" when rate limiting is operating in degraded
mode (Redis unavailable). Values in other rate limit headers are
best-effort estimates.
schema:
type: string
enum:
- 'true'
responses:
ErrorResponse:
description: Error response
content:
application/json:
schema:
$ref: '#/components/schemas/HttpError'
RateLimitedResponse:
description: >-
Too many requests. Short-window RATE_LIMITED responses include retry
timing and rate limit headers. Monthly trial quota responses use
MONTHLY_TRIAL_QUOTA_EXCEEDED with retryable: false.
headers:
Retry-After:
$ref: '#/components/headers/Retry-After'
X-RateLimit-Bucket:
$ref: '#/components/headers/X-RateLimit-Bucket'
X-RateLimit-Limit:
$ref: '#/components/headers/X-RateLimit-Limit'
X-RateLimit-Remaining:
$ref: '#/components/headers/X-RateLimit-Remaining'
X-RateLimit-Reset:
$ref: '#/components/headers/X-RateLimit-Reset'
X-RateLimit-Degraded:
$ref: '#/components/headers/X-RateLimit-Degraded'
content:
application/json:
schema:
$ref: '#/components/schemas/HttpError'
securitySchemes:
bearerAuth:
type: http
scheme: bearer
bearerFormat: API key
description: 'Use Authorization: Bearer sk_test_* or Authorization: Bearer sk_live_*.'
````