Your Deployments Are Lying. AI Knows
summary
CI/CD shows green while 10% of users see checkout or cart failures; logs show slight wording shifts and one trace span slows by ~150 ms. The demonstration ties Prometheus + Pyod + Grafana + Open Search + Langchain + Jerger + OpenTelemetry collector + captain into a pipeline that runs post-deploy anomaly detection, semantic log summarization and behavioral baselining, surfaces correlated alerts, and reduces MTTD/MTTR in practice. behavioral baselining semantic log summaries
AI-written summary · about
chapters (11)
- 0:00Deployment reliability issues
- 3:14Green deployments' false security
- 5:19Business failures missed by monitoring
- 7:26Static thresholds limitations
- 9:55Intelligent observability needed
- 11:19AI anomaly detection
- 12:47Traditional vs AI monitoring
- 14:17AI in CI/CD pipelines
- 16:03AI log and trace analysis
- 17:09Key takeaways and solutions
- 17:54Q&A
related talks
28mEnd-to-end CI/CD observability with open source
44mAgentic Delivery, Guardrailed: AI in CI/CD & Platform Engineering (Open Source Patterns)
26mIntelligent Deployment Pipelines
34mDeploy, Release, CI/CD, Oh My! DevOps for the Rest of Us
29mFrom Flaky Builds to Self-Healing Deployments: An OTel-Driven CI/CD...
35mGaining Observability Into Your CI/CD Pipeline
43mTo make mistakes is human. To auto-deploy it is DevOps.
38mEnhanced Log Analytics Using Generative AI