TonyAlicea10

I am in what I feel is a unique position: I’ve been a dev and dev educator for 25 years, and I also have spent 15 years in HCI/UX/User Research.

I’ve spent countless hours facilitating usability tests and then building the software in response to them.

I 100% agree with the “dribblization” of usability engineering.

More so, I think the term “design” has become overloaded. A graphic designer who wants to actually get paid and eat goes into UX. But where does their joy come from? From the visual, not the usability.

This year I got hired at Nielsen Norman Group, helping train designers to understand the theories of the human mind and applying them to UI.

I am finding the range of knowledge and background amongst designers is massive, precisely because of the imprecision of the word “designer”.

If we called it “applied cognitive and behavioral psychology” we would get closer to the results we want and perhaps a smidge more respect for its necessity in the age of slop UI.

Yes those are cool ideas! The push and pull of ideas has been between satire and playability. But skill atrophy is definitely a game mechanic I've been exploring.

There's a probability per-level of mutation.

I've been thinking a lot about how agentic coding is similar to gaming. Risk, reward, gambling, gratification, etc. I've been exploring that by vibe coding some satirical games that aim to reflect what agentic coding can be like in a deadline environment. This is one of those games.

It's Tetris but:

- You have to build a tetromino either manually or have AI "generate" it.

- Generated tetrominoes might mutate after falling.

- You can spend time to "review" generated tetrominoes as they fall to see how they will mutate.

The theme is the risk/reward/pressure of agentic code vs manual vs reviewed agentic code in a deadline environment.

I like to say AI is like the super soldier serum from the MCU. Teams take it and either become Captain America or Red Skull, it depends on what you were to begin with.

The fact that it’s called “the quick pitch” screams Claude-written pithyness pulled from some context that doesn’t match the article’s style (like investment pitch decks).

We don't run our own unit tests on compiled code because it generally can be trusted. LLMs are an extraordinarily leaky abstraction.

Saying "as long as it works and tests pass" suggests that we can test for every possible scenario. We can't. And tests can be flawed on top of it. Which is why an LLM is no more a compiler than a human coder (as you say) is.

I think two things contribute to the trust in LLMs:

1. Decades of generally reliable systems (I trust my calculator because it always says 1+1=2) has trained people to believe what computers say.

2. LLMs "speak" with great confidence, which has, as you say, a psychological effect.

It absolutely might make mistakes. But the skill requires all questions to be based on lines of code with specific line numbers which the human needs to review. That's intentional to serve as a safeguard against hallucinations.

I mean in the traditional software sense. We trust autopilots to fly planes. Would you get on a plane flown by an LLM?