How to Ship AI-Generated Code to Production
AI writes code. But shipping to production? That still takes a software engineer.
In this GitKon talk, Chris Kelly from Augment Code breaks down what it actually means to use AI-assisted development to write production-ready code, not vibe code. If you've been using AI coding assistants and wondering why the output doesn't always make it past code review, this is for you.
Chris covers:
00:00 - Why "AI is replacing devs" headlines miss the point
02:45 - Code is the artifact, not the job
05:10 - Why LLMs can't make architectural trade-offs (and you still can)
08:30 - The airplane metaphor: fixing software that's already flying
11:00 - How to give AI agents the right frame to work in
14:15 - Why context is everything in AI prompting
17:40 - Code review as the most important skill of the AI era
21:00 - How Augment Code's context engine changes the game
Key takeaway: The engineers who will thrive aren't the ones who let AI do everything. They're the ones who know how to review, direct, and architect around what AI produces.
#GitKon #AIcoding #SoftwareEngineering
GitKraken Desktop:
http://tr.ee/GKDYT
GitKraken CLI:
http://tr.ee/CLIYT
GitLens for VS Code:
http://tr.ee/GLYT
Git Integration for Jira:
http://tr.ee/GijYT
Git Blog:
http://gitkraken.com/blog