Apps, Stacks, and Frameworks: Avoiding “Shiny Object” Syndrome
summary
A team replaces a relational model with MongoDB to ship a new UI, finishes early, then sees traffic degrade when 15 million documents expose terrible queries and wrong indexes; recovery comes from moving low‑use rows into cold storage and the outage proves the prototype didn’t scale. The concrete failures are lack of no load testing, unclear operational runbooks, and missing monitoring — vet performance claims, simulate production volume, and codify SOPs before deployment.
AI-written summary · about
chapters (8)
- 0:00Introduction and background
- 1:47Early career and DevOps roots
- 5:08The MongoDB case study
- 9:17Post-deployment challenges
- 13:03Lessons in technology adoption
- 18:10Management and leadership
- 23:32Individual contributions
- 28:16Conclusion and final thoughts
related talks
36mApps, stacks & frameworks: Avoiding "Shiny Object" Syndrome
24mApps, Stacks, and Frameworks
26mArchitecting for scale
20mDoing Things the Hard Way
55mShiny objects are cool, but so is building products people use
27mParadigm Lost
46mDigital scale-up architectural patterns
41mThe monolith must die! - How to convince managers