The Demo Is Easy. Production Is Hard.
summary
Vibe Coding makes prototypes fast but production requires discipline: build a schema-first data model, prevent database drift with versioned migrations, and enforce CI/CD and automated security checks. Plan fault resilience, BCDR and support rotations, decide single-tenant rollout vs multi-tenant early, and keep deterministic paths when money or audits matter (LLMs aren’t). Ship a small harness that continuously tests API keys and endpoints, and use AI-generated technical docs and changelogs so audits and operators can reproduce and restore state.
AI-written summary · about
chapters (10)
- 0:00Live coding popularity
- 3:20Demos crack in enterprise
- 7:55Database drift in vibe coding
- 12:13Enterprise IT problems
- 16:55Six capabilities for trust
- 20:13Trust as product feature
- 23:23Production discipline
- 24:33Crisis management
- 28:51Developer steps to avoid risk
- 32:53Closing remarks
related talks
49mVibe Coding for Production
48mHow to deliver continuously
30mProduction › Prototype: How a Small Team Replaced a Core Platform in 90 Days
29mClosing the Loop: How GraphQL Gives Coding Agents Eyes on What Actually Matters
37mReal Developers Ship (a.k.a Tenets for Software Delivery)
34mHow to bridge the gap between Data Scientist & Decision Makers?
29mLeveling Up from Planning to Production
53m5 tips on how to deliver software effectively by aligning coding and business best practices