Technical Enshittification: Why Everything in IT is Horrible Right Now and How to Fix It
summary
Argues that modern software rot is driven by context degradation—reorgs, bloat, slow CI and AI-generated code that add tech debt and vulnerabilities—illustrated by Sonos’ 1.1 app rating and the 2024 CrowdStrike update that grounded major airlines while Southwest ran legacy systems. Cites surprising metrics: average build roundtrip ~25 minutes, developers write code ~52 minutes/day, ~1M malicious packages and 60% of orgs report serious tech debt. Recommends concrete fixes: spec-driven development, intent integrity chain, and using agent-backed tools and Tessle tiles to keep specs, tests and AI in sync.
AI-written summary · about
chapters (9)
- 0:00Technical enshittification intro
- 3:51Catastrophic software failures
- 10:16Bloat and risk escalation
- 17:17Lost context costs
- 23:01Context hunting impact
- 29:48AI’s role in context loss
- 34:01Preserving context strategies
- 41:53Summary and call to action
- 42:54Q&A on docs and AI
related talks
53mTechnical Enshittification: Why Everything is Horrible and How to Fix It
01h 02mTech trends in focus: A fireside chat
28mTechnical Debt: Why it'll ruin your software
11mTechnical Debt Crisis! When Non-Engineers Write Code
39mStop Focusing on Tech Debt
42mThe Future of Software Development
01h 07mTechnical Neglect
36mNo More ‘Someone Else’s Problem’: Tackling Tech Debt at Scale