Beyond CodeGen: How AI & Machine Learning Are Transforming DevOps
summary
Claim: generative code is only a slice and AI/ML is instead reshaping delivery platforms to cut developer toil across the SDLC. Examples: using Harness as one master pipeline to eliminate 5,000-pipeline sprawl, telemetry and observability for instant rollbacks or rapid roll‑forwards, feature flags and canaries, and FinOps/cloud cost management pushed left. Surprising numbers: 100 million in one year (OpenAI) versus Netflix’s decade, and Netflix gave up counting at 4,000 deployments a day. Emphasis: guardrails over gates to preserve speed with safety.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:26Netflix to OpenAI scaling
- 3:11Move fast and break things
- 5:16Guardrails over gates
- 8:08Tier one vs tier two platforms
- 10:37Business dependency on deployments
- 12:245,000 runways in pipelines
- 14:28Journey versus destination
- 16:53Deployment speed and reliability
- 18:38Roll back vs roll forward
- 20:52Measuring DevOps success
- 23:14KPIs and business metrics
- 27:04Scaling stages and metrics
- 29:13Cloud cost visibility left
- 32:48Continuous delivery investment
related talks
38mGenerative DevOps
36mGenerative DevOps
39mDevOps in the Age of AI: Human Powered Evolution
28mAIOPS the power of DevOps at scale
27mTransforming Developer Platforms with Generative AI
38mFrom code to culture: How AI is reshaping the engineering organization
36mDeveloper Experience Trends & Predictions
20mGenerative Value Streams: When Code is not a Constraint