RSS Feed Subscribe to RSS Feed


Blog post summary: Shipping fast and safe by Kesha Mykhailov at Intercom

I like this “Shipping fast and safe: Building a culture of low-risk learning” article by Kesha Mykhailov at Intercom.

Some highlights…

There’s only so much you can do with automated tests and pre-production environments. Production is the only place where your code, infrastructure, and customers come together so that you can truly validate your code, and learn how your customers use your product.

Rather than creating artificial barriers to shipping, minimize the risk of the “unknown unknowns” by:

  • Encouraging engineers to ship the smallest thing as quickly as possible
  • Continuously invest in the delivery pipeline & observability
  • Reducing the blast radius of potential issues
  • Have fast recovery times

But don’t just ship fast, ship safely:

  • Be available after shipping (your in-depth knowledge of the change may allow you to spot problems faster than alerts)
  • Ship instrumentation first and code second
  • Use feature flags (minimize the blast radius; ease rollbacks, but avoid leaving too many feature flags hanging around)
  • Ship to a small subset of the traffic
  • Ship the “read” path first


Tags: , ,

Leave a Reply