How to be quick without being dirty: maintaining quality at high velocity
summary
Practical, checklist-style guidance for speeding up without exploding quality: use Shipyard ephemeral environments and production-like test data, capture decisions with ADRs and quick Mermaid diagrams (Notion/Excalibur renderers), follow the rule write a test after it breaks, keep CI/CD and feature flags ready as kill switches, isolate coding agents via config files and sandboxes, and measure with DORA/SPACE plus opposing “tension metrics” so speed and failure rate stay balanced.
AI-written summary · about
chapters (9)
- 0:00Introduction to velocity and developer productivity
- 1:41Planning framework and avoiding overengineering
- 2:51Making smart architectural decisions and reducing risk
- 5:37Relentless prioritization and managing technical debt
- 7:20Implementing guardrails and testing strategies
- 8:43Establishing safety nets via logging and monitoring
- 10:46Best practices for coding and git hygiene
- 12:38Framework for managing AI coding agents
- 14:48Measuring developer productivity and performance
related talks
38mBuilding a high quality product
21mNeed for speed - Moving fast as a software engineer
53mNever-Neverland of pursuing Quality
31mBalancing New Features While Maintaining Quality
35mBetter, Faster, Stronger - Delivering High Quality Products
28mHow to prioritize dev experience in the development process
58mTechnical Practices that Take Time but Save Time
30m10 Ways to Improve How You Ship Software