Photoshop

So we're optimizing LLM token consumption now by... using external stylesheets? The same practice we've been preaching since 2005? Incredible. The AI era has brought us full circle to basic web development best practices, except now the justification is "save tokens" instead of "save bandwidth." The beauty here is watching people discover that separating concerns actually has benefits beyond making your code maintainable. Who knew that not dumping 20 lines of CSS into every prompt would reduce token usage? Next you'll tell me that minifying code and using compression also helps. The real galaxy brain move is training the LLM to reference external CSS so it "never outputs CSS again." Because nothing says efficiency like teaching an AI to avoid generating something it's perfectly capable of generating. It's like hiring a chef and then telling them to never cook vegetables because you bought them pre-cut.

Nothing says "job security" quite like being the only person who knows where the bodies are buried in production. Got replaced by an AI SWE? Cool, cool. Now you're a contractor charging triple your old rate because guess who they're calling at 3 AM when everything's on fire? The beautiful irony here is that companies think they're saving money by replacing engineers with AI, right up until they realize the AI doesn't know why the deployment script needs that one random sleep(5) command or which environment variable typo is actually load-bearing. Suddenly that $200/hour contractor rate doesn't seem so bad when your entire revenue stream is returning 500s. Pro tip: Always leave at least one undocumented quirk in your infrastructure. It's not job security, it's "knowledge transfer optimization."

The duality of modern developers: initially impressed by shiny new AI features, followed immediately by the desperate search for the "off" button. GitHub Copilot might finish your code, but it also finishes your patience when it confidently suggests the wrong solution for the fifth time in a row. The real 10x developer move isn't using AI—it's knowing exactly when to turn it off.

Ah, the ultimate job security plan! Turns out the one person AI won't replace is the colleague who contributes absolutely nothing. That sweet spot where you're so useless that even automation can't justify the ROI of replacing you. It's like finding a bug so bizarre that QA just labels it as "expected behavior" and moves on. The perfect defense against technological unemployment isn't upskilling—it's strategic incompetence!

Content Al model when it sees data it wasn't trained on

When your portfolio claims "full stack web app with backend" but the entire backend is literally just two Express routes copy-pasted from Stack Overflow and a JSON file pretending to be a database. Sure, it technically has a backend... in the same way a cardboard cutout technically has depth. The "No AI" disclaimer is the cherry on top—gotta make sure everyone knows you typed those two commits yourself, even if one of them was just fixing a typo in the README.