Three pillars of observability applied to ML.
mediumAnswer
- (1) Metrics: aggregated numeric signals over time (Prometheus).
- ML: prediction distribution, latency, drift score, feature freshness.
- (2) Logs: structured events with context (Loki, ELK).
- ML: prediction logs with input + output + version + latency.
- (3) Traces: causal chain of a single request across services (Jaeger, Tempo, Datadog APM).
- ML: feature fetch → preprocess → model call → postprocess.
- Modern add: profiles (continuous profiling).
- Enable debugging beyond dashboards.
Check yourself — multiple choice
- Random
- Metrics (Prometheus): aggregated numeric; Logs (Loki/ELK): structured events; Traces (Jaeger/Tempo): causal chain per request; ML adds prediction/feature/latency + traces per request; profiles bonus
- Just logs
- Not real
Observability: metrics + logs + traces + profiles; each for different debug.
#observability#monitoring
Practise MLOps & Data Quality
215 interview questions in this topic.