Fast & Simple Deployments for Code & Infra
summary
Shows how Honeycomb uses CircleCI, Terraform (plus Chef and home‑grown scripts) to push green builds and deploy roughly a dozen times a day with ~3‑hour lead time. Code emits events to Honeycomb for observability, tests and Storybook Golden Masters catch UI regressions, human PR reviews combine with automated CI, and feature flags separate deploy from release. Infrastructure is gitops’d, migrating toward Kubernetes, with separate dog food/kibble environments and separate data stores; a 10‑minute ingest outage burns 70 days of a 99.9% SLO.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 2:13Engineering constraints and goals
- 9:20Observability and feature flags
- 18:53DevOps metrics and deployment frequency
- 23:04Infrastructure automation and Kubernetes
- 31:56Service level objectives and error budgets
- 37:59Balancing stability and features
- 39:23Q&A
- 41:50Q&A
- 43:16Q&A
- 44:59Q&A
- 46:22Q&A
- 48:10Q&A
- 49:55Q&A
- 55:57Closing remarks
related talks
35mBeyond build/test/deploy
42mTrick or Treat Your Pipeline: A Haunting Tale...
01h 03mYet Another Continuous Delivery Talk
14mProgressive Delivery with Observability
36mContinuous Deployment with Rails
06mIgnite - Blue Green Deployments Using Terraform + Packer + Jenkins
49mAutomation in Deployment on Hybrid Hosting and Private Cloud Environments
25mIncrease App Confidence Using CI/CD and Infrastructure as Code