Hanzo O11y
Full-stack observability platform
Hanzo O11y
You cannot fix what you cannot see. Hanzo O11y is how you watch a running system — every metric, log, and trace in one place, with dashboards and alerts that surface trouble the moment it starts. Built on Prometheus, Grafana, OpenTelemetry, and Loki.
Highlights
- Prometheus Metrics -- 15-second scrape interval, 30-day retention, PromQL
- Grafana Dashboards -- Pre-built dashboards for every layer of the stack
- Distributed Tracing -- OpenTelemetry-native with automatic context propagation
- Log Aggregation -- Structured log ingestion and LogQL queries via Loki
- Alerting -- Threshold, anomaly, and SLO-burn-rate alerts to PagerDuty, Slack, webhooks
- SLO Management -- Define, track, and alert on Service Level Objectives
Endpoint: o11y.hanzo.ai
Prometheus: o11y.hanzo.ai:9090
In this section
Metrics
Product metrics, events, and sessions — OpenTelemetry- and Prometheus-compatible time series for your…
Logs
Structured logs across all services — OpenTelemetry-compatible, correlated to traces by trace id.
Traces
Distributed traces across all services — OpenTelemetry-compatible, correlated with logs, metrics, and scores.
Sessions
Traces grouped into multi-turn sessions — one conversation or agent run across many requests.
Dashboards
Product analytics and observability dashboards for Hanzo Cloud -- usage, cost, latency, and health in one…
Datasets
Curate evaluation datasets and items — the input/expected-output pairs your experiments run against.
Experiments
Dataset runs, comparisons, and experiment analytics — score a dataset against a model and compare runs.
Scores
Evaluation scores from feedback, graders, and review — attached to traces, observations, and sessions.
Score Configs
Score definitions — data types, ranges, and categories that every score conforms to.
Annotation Queues
Review queues for scoring traces and observations against a defined set of score configs.
How is this guide?