Hangar DX Podcast Fearless deployments with Charity Majors
Jun 16, 2025
Charity Majors, Co-founder and CTO of Honeycomb.io, shares insights on fostering a culture of continuous deployment in software engineering. She emphasizes the importance of viewing production as a dynamic system and the need for robust feedback loops. The conversation highlights the evolving role of Site Reliability Engineers and the significance of a blameless culture. Charity also discusses AI's potential to enhance observability, advocating for transparency and training to streamline software delivery.
AI Snips
Chapters
Transcript
Episode notes
Make Shipping Routine and Boring
- Make shipping code a regular, consistent, and expected activity to reduce deployment risks.
- Ship so often that deployments become routine and boring, which is the safest state for a company.
Prerequisites for Continuous Deployment
- Trust your tests and keep build and deployment times short to enable continuous deployment.
- Avoid large batch deploys as they increase risk; smaller, frequent deploys reduce cognitive load and improve safety.
Feature Flags Empower Safer Deployments
- Use feature flags to decouple deployment from release, enabling safer, reversible changes.
- Combine feature flags with good observability and progressive deployments like canaries to minimize production risk.