Designing Holistic Security Programs
summary
Argues that moving security left and layering imperfect controls wins: design-time controls save massive late-stage costs and prevent a risk of a $200 million exploit. Recommends defense-in-depth, separation of duties and least privilege, not performative bounties. Gives concrete builds: a Solana C client with per-tile sandboxes that mutes memory-safety/RCE risk, and a Wormhole-style rate-limited governor that caps cross-chain outflow and buys response time — both described as design-once mitigations that actually work.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:42Be Boring
- 3:55Enduring Principles
- 7:24Security Smells
- 10:09Outsourcing
- 11:13Defense Depth
- 14:45Conclusion
