AI's Impact on Development, Post-Production & Support Journeys
summary
Survey of concrete AI features and limits for developers and ops: AI assistants like GitHub Copilot, Codium and Codi produce a readable summary of legacy files often good for 70% gist, generate clean recursive utilities, and find every repo touchpoint when adding a DB field. Automated unit tests hit 90–95% coverage on many modules but require human-written tests for validation. Tools surface code smells, flag vulnerable package versions, suggest a circuit breaker for sync calls, extract the exact log snippets for alerts, and auto-assign past responders; Atlassian is shipping related product features.
AI-written summary · about
chapters (12)
- 0:00Introduction to AI tools
- 1:15Legacy code understanding
- 2:38Code utility generation
- 3:41Database change impact
- 5:03Unit testing automation
- 6:48AI-assisted code review
- 7:52Vulnerability detection
- 8:54Scalability testing insights
- 9:57Industry trends in AI
- 10:59Post-production AI support
- 12:04Log analysis automation
- 13:08Incident response assignment
related talks
14mDeveloper productivity in a post-AI world
27mAI for software development: A reality check
12mThe Synergistic Impact of Artificial Intelligence on DevOps
24mDevelopers and AI Enemy or Frenemy Leveraging AI and Atlassian to empower developers
44mAI and Code Quality: Building a Synergy with Human Intelligence
24mAI and Web Development: Hype or Reality
57mOne Team’s Journey to Using AI Without the Hype
01h 16mAI in Software Development: Boosting Efficiency