Production Engineering When Trading Billions of Dollars a Day [video]

· Source ↗

TLDR

  • Jane Street production engineer Mark Dos walks through daily trading ops, incident response, and why SLO-based monitoring often falls short at billion-dollar scale.

Key Takeaways

  • Every message matters when software has near-unlimited bank account access; alerts and incident response have direct, measurable P&L impact.
  • Traditional SLO-based monitoring approaches are often insufficient in trading environments due to the stakes and real-time nature of markets.
  • Defense in depth and cross-team communication are core pillars of Jane Street’s production engineering model.
  • Talk covers trading-environment-specific characteristics that make production engineering uniquely high-risk compared to standard ops roles.
  • Concrete case studies illustrate how Jane Street handles incidents across global stock markets.

Hacker News Comment Review

  • No substantive HN discussion yet.

Original | Discuss on HN