Platform health · Live

System status

All systems operational. This page reports the live availability, uptime, and response time of every AI Agentics service — the API, agent runtime, dashboard, integrations, webhooks, and distributed tracing — so you always know the platform is healthy before you ship.

  • All operational
  • Updated 2026-06-20
  • 99.99% uptime (90d)

Every component below is monitored continuously from multiple regions with synthetic agent runs and real-traffic health checks. When a check fails, we open an incident here within minutes and post updates until it is resolved. The current AI Agentics status is operational across all services, with rolling 90-day uptime tracked per service against our published SLA targets.

If you are debugging an integration, start with the service tiles below, then check the response-time trend and the incident history. For deeper diagnostics, our documentation covers retries, timeouts, and idempotency, and the security page documents our SOC 2 controls and data handling.

Live services

Service status at a glance

Each service is health-checked every 30 seconds. Green means requests are flowing within target latency and error budgets.

REST & streaming API

Operational

The core HTTPS API for creating runs, calling tools, and streaming tokens. p95 latency within target; zero elevated error rate.

Agent runtime

Operational

The execution loop that plans, calls tools, and orchestrates multi-agent runs. All worker pools healthy with spare capacity.

Dashboard

Operational

The web console for building agents, inspecting runs, and managing keys. Static assets served from the global edge.

Integrations

Operational

Connectors to vector stores, databases, Slack, and third-party APIs. All upstream providers reachable.

Webhooks

Operational

Outbound event delivery for run lifecycle and tool callbacks. Delivery queue drained; signed payloads verified.

Tracing & observability

Operational

Per-step traces of every decision, token, and tool call. Ingestion and search are real-time with no backlog.

Live metrics

Uptime and response time

Rolling availability for the trailing 90 days, plus median API response time over the last eight days.

99.99%

API uptime (90 days)

Target SLA: 99.9%

Median API response time (ms)

1314151617181920
Daily p50 latency over the last eight days. Lower is better; all values are well within the 400 ms target.
At a glance

Reliability summary

99.99%

Uptime

trailing 90 days

171ms

Avg API latency

p50, last 24h

0

Active incidents

right now

2

Incidents

this quarter

Incident history

Recent incidents

A transparent log of past disruptions. All resolved, with root-cause notes and the safeguards we added.

  1. 2026-05-28Resolved

    Elevated webhook delivery latency

    A queue worker deploy reduced delivery throughput for ~22 minutes, delaying outbound webhooks. No events were lost; the backlog drained automatically after rollback. We added per-worker saturation alerts.

  2. 2026-04-09Resolved

    Degraded API latency in one region

    A connection-pool exhaustion bug pushed p95 latency above target in us-east for 14 minutes. Traffic was shifted to a healthy region; a fix and a tighter pool ceiling shipped the same day.

  3. 2026-02-17Resolved

    Tracing ingestion delay

    Trace search lagged by up to 3 minutes during an indexing migration. Run execution was unaffected. We now run migrations behind a dual-write path to keep ingestion real-time.

Get notified about incidents

Want a heads-up the moment a service degrades? Subscribe to status updates and receive email alerts when an incident is opened, updated, or resolved — plus advance notice of scheduled maintenance. Reach us via the contact page to join the list or set up a webhook for your on-call channel.

API referenceSecurity & SOC 2ChangelogRoadmapuptimeincident historyresponse time
Build with confidence

Ship agents on a platform you can trust

99.99% uptime, real-time tracing, and SOC 2 controls — so your agents stay up while you focus on building.