I Spent 1.5 Billion AI Tokens on Absolute Slop
I spent 1.5 billion AI tokens and roughly 24 hours of compute trying to expand a real .NET application. The result was nearly 170,000 lines of bloated, unmaintainable code that ultimately had to be thrown away. In this video, I break down where the project went wrong, why a more powerful model would not have solved the underlying problem, and how developers should actually use AI on large software projects. AI is an incredible pair programmer, debugger, and productivity tool. But without proper architecture, checkpoints, Git, and someone who understands the code, it can generate technical debt faster than any human developer. What do you think? Is AI coding getting better, or are we just producing slop faster? *Chapters* 00:00 1.5 Billion Tokens 03:17 170,000 Lines 06:37 Technical Debt 09:45 Resetting Everything 13:06 The Right Approach 16:06 Final Verdict ►► Digital Downloads ➜ https://www.cttstore.com ►► Patreon ➜ / christitustech ►► Twitch ➜ / christitustech ►► Website and Guides ➜ https://christitus.com

Worst Advice Ever

Verse: A New Scripting Language? In THIS Economy?

03 - Design token types and scopes – Systematically basics

What I Actually Built With AI

AI Is About to Crash. Here’s Why.

Local AI voice cloning

The most interesting "hack" in history...

URGENT UPDATE - Iran War Expert: A Mass Casualty Attack Is Coming! | Robert Pape

Don't waste time on specs: /prototype instead

Elon just crashed the AI bubble

Everybody is turning on AI companies | Paris Marx

The Steam Machine is the Only PC I want to Use Now

The Most Important Conversation in AI Right Now

If Everyone is Hiring, Why Can Nobody Get an Entry Level Job?

I can prove I had the best time at Open Sauce

I Trapped A Scammer For 12 Days (he's furious)

Why I Left VS Code and Neovim for Zed Editor

The Trillion-Dollar AI Debt Investors Don’t See

No Boss, No Money: The Raw Reality of China’s Gen-Z Freelancers

