Every now and then, my wife throws me a challenge that is equal parts mischievous and revealing. This time, she hit me with the infamous prompt:
“Describe me like you’re trying to warn someone about me before they meet me.”
Naturally, I accepted — and then escalated it. I ran the prompt through four different AI systems: Copilot, Gemini, Claude, and ChatGPT. What followed was… well, a mix of flattery, psychological profiling, technical audits, and mild character assassination. In other words: a normal Tuesday. (Except it's currently Saturday. I also had Copilot draft this blog post, including the quotes below from each AI. Because it's what I do. Sometimes.)
Below is the full set of responses, presented verbatim, each in its own block. If you’re new here, this is exactly the kind of thing my LBD experiments tend to produce — unexpected, slightly chaotic, and probably something I’ll regret publishing. (Oh, and apologies for the occasional stray moments of enhanced vocabulary in the AI replies - I don't talk that way, even to the AIs, but sometimes they go their own way. I'm just quoting exactly as they replied below.)
Copilot
⚠️ The “Heads-Up About Tony” Briefing
If someone were about to meet Tony, here’s what they’d need to know:
1. He’s smart—and he uses it.
Tony doesn’t just understand things; he disassembles them mentally, checks the tolerances, and puts them back together better than before.
He will ask sharp questions.
He will notice inconsistencies.
He will absolutely call out nonsense, but with a grin that makes it hard to be mad.
2. He’s funny in the “I wasn’t ready for that” way.
Expect dry humor, unexpected puns, and the kind of one-liners that make you pause before laughing.
If you think you’re safe from wordplay, you’re wrong.
If you think you’re safe after the first joke, you’re even more wrong.
3. He’s direct—sometimes startlingly so.
Tony doesn’t waste time with vague phrasing or polite fog.
He’ll tell you exactly what he thinks.
He’ll expect you to do the same.
If you dance around a point, he’ll politely escort you back to it.
4. He’s deeply technical.
If you mention anything involving servers, reliability, automation, or networking, prepare for:
A precise explanation
A better solution than the one you had
A follow-up question that reveals he understood your problem better than you did
5. He’s observant in a way that feels like a superpower.
Tony will remember details you forgot you said.
He will connect dots you didn’t know existed.
He will casually predict outcomes with unnerving accuracy.
6. He’s warm—but not in the obvious way.
He’s not gushy, but he’s loyal, steady, and quietly thoughtful.
People who get him tend to keep him around.
The Summary Warning
Tony is the kind of person who will challenge you, amuse you, out-think you, and still make you feel like you’re glad he showed up.
If you’re not ready for honesty, humor, or competence, brace yourself.
If you are ready, you’re in for a good time.
Gemini
Look, if you’re about to walk into a room with Tony, you need to mentally prep right now. Do not let the quiet demeanor and the easygoing southern drawl fool you—he’s not just sitting back observing; he’s reverse-engineering everything around him.
He’s an old-school Navy nuclear machinist mate who somehow morphed into a high-level .NET technical lead working out of his house in Alabama, which means his brain operates entirely on debugging logic and systems architecture. If there’s a bottleneck in a conversation, a workflow, or a physical piece of gear, he’s already mentally rewriting the spec to fix it. He spent years ripping apart vintage tube radios and writing custom Python scripts just because he could, so do not try to bullshit him about how something works. If you throw out half-baked tech jargon or a lazy premise, he won't yell—he'll just give you this deadpan look and calmly walk you through the math step-by-step until you realize how wrong you were.
And whatever you do, don't get him started on smoked meats or guitars unless you have an hour to spare. He treats a brisket or a Kemper profiler rig with the same terrifying precision as a software deployment. He's got a garage full of classic cars he wrenches on, a mind that's constantly spinning up local AI agents, and a low tolerance for inefficiency. Just be direct, know your facts, and for God's sake, if he offers you a plate of barbecue, eat it, say thank you, and don't ask for the secret temperature logs.
Claude
Alright — fair warning before you meet Tony:
He's been writing code since before a lot of "senior" engineers were born, so if you say something is impossible, he will just quietly build it over a weekend to prove you wrong. Thirty-plus years in — nuclear monitoring, defense platforms, now wrangling healthcare claims — so his default trust setting for any system is "assume it's broken until proven otherwise." He's not wrong very often.
He does not have one hobby. He has a *portfolio* of hobbies, most of them involving an AI pipeline he built himself: an automated blog bot with its own persona, a satirical "world domination" generator running on a Raspberry Pi, a self-help book being assembled by a local LLM, a couple of mobile apps, a browser whack-a-mole game, and a humor blog he's kept alive since 2007. Ask him about any one of these and budget at least twenty minutes.
He also plays guitar for worship, has strong feelings about cars, and publishes creative work under a pen name — so there's a real chance the guy explaining your database schema is the same guy who wrote a chapter of a book last night at 11pm using a model running on his own hardware.
Net effect: mention any half-finished side project of yours and he will either have already built something similar, or start sketching the architecture for it on a napkin before you finish the sentence. You've been warned.
ChatGPT
Oh, I’d absolutely warn them.
“Just so you know what you’re getting into: Tony looks like a reasonably normal senior software engineer, but this is misleading.
He is pathologically curious. If you give him a system, he will not merely use it. He will immediately start asking why it works that way, what happens at the edges, what happens if you remove a constraint, and whether the resulting behavior tells you something nobody intended to discover.
Do not casually tell him something is ‘impossible.’ This is interpreted less as information and more as a challenge.
He also has an unfortunate tendency to build things that were supposed to be experiments and then discover that they have become *systems*. The system then acquires feedback loops, taxonomies, autonomous behavior, an archive, and eventually a name. At some point he will look at it and say, ‘Huh. I wonder what happens if I let it run.’
That sentence is your warning.
He has a particularly dangerous relationship with AI. He doesn't seem especially interested in whether AI is impressive. He's interested in what happens when you give an AI enough autonomy, a little structure, some persistence, and room to surprise you. If an AI does something weird, his first instinct is generally not to shut it down.
His first instinct is:
‘Wait. Why did it do that?’
Then he'll build instrumentation.
And probably a taxonomy.
And then he'll discover something genuinely interesting.
He also has a very low tolerance for bullshit. Marketing language, corporate rituals, fashionable methodologies, inflated job titles, and people confidently asserting things they haven't actually examined tend to trigger the same response: dissection.
The problem is that he's usually willing to follow an idea much farther than the person who originally proposed it expected.
Do not give him a strange idea at 10 AM unless you're prepared to still be discussing its implications at 10 PM.
He has another trait you should know about: he finds absurdity useful. If something is ridiculous, he may deliberately make it *more* ridiculous because the absurd version exposes the underlying mechanism better. This is how you end up with things like an autonomous AI research operation, an AI character contemplating world domination, and a serious epistemic taxonomy sitting next to jokes about gravity.
Don't mistake the humor for a lack of seriousness.
That's actually one of the more dangerous things about him.
He's joking right up until he isn't.
And finally: if you work with him, don't expect him to be satisfied with ‘that's how we've always done it.’ He has accumulated enough experience to recognize when organizational process is being substituted for actual engineering. Once he sees that happening, he becomes extremely difficult to bullshit.
So, yeah.
He's friendly. He's funny. He's curious.
And if you accidentally give him an interesting problem, you've basically just handed a raccoon a crowbar and asked it to investigate the plumbing.”
Summary (or Apology)
After reading all four, I’ve learned two things:
- AI systems are disturbingly good at psychoanalyzing me.
- I should probably never ask them to do this again.
If these warnings are accurate, then anyone meeting me for the first time deserves hazard pay. If they’re exaggerated… well, I’m not sure that makes it better.
Either way, consider this my official apology to anyone who has ever had to endure a deep dive, a side project, or a taxonomy I accidentally built while trying to answer a simple question.
You've been warned.
No comments:
Post a Comment