MrDunham

I think it's brilliant. I'm now paying $30 per month for Github actions as a duo shop... it's freemium working the way it should. Give them the first hit and all.

And yes, I need to optimize (and am), but they're already getting hundreds from little me. I even started going towards the CI doing browser walks, code tests, AI rework, etc. It saved me from having another dev machine.

Great stuff, powerful as all heck, and once you see the power it was—for me—impossible NOT to think of 6 other uses and blow through the free CI tier.

I think this is a very important point. I've specifically started thinking about my AIs as humans. Not in the anthropomorphized sense, more like NOT treating them as deterministic software.

The challenge I've had is I tell it to do {thing}, it does {otherThing} after getting distracted. Then I get annoyed (let's not mention how I probably half-assed the instructions and wouldn't expect a senior human to be able to succeed).

For me, it was a CI/CD issue with our two person startup. I bypass CI/CD often because it was built to catch the AIs. Damn thing went chasing rabbits. Later that day I talk to my cofounder, who says: I have to go chase down this very important CI/CD issue!

Turns out both humans and AIs get distracted relatively easily.

I find that if I think of my AIs less like (deterministic) software and more like leading actually employees that I both get better output and curse (a lot) less.

LLMs _were_ trained on human writing, so it makes sense to me that they tend to act human-like... for better and worse. So yeah, they do dumb stuff, and so do people.

Woah... thank you for sharing RE the Pencil, because that was the only killer feature that I could think of that would get me to buy the Duo. Taking notes during an in-person meeting with the pencil feels (and looks) much more "they care about what I'm saying" VS "they're texting/emailing someone" (when typing on a phone or computer).

And even with mobile dictation/notetakers, there's something that screams respect when someone writes actual notes down by hand.

Big big deal indeed, thanks for sharing.

It landed back where it started.

Seems very appropos.

The iPhone camera bump is the "jumped the shark" moment for me when Apple went from unwilling to accept that level of quality to "I'm not sure... they might". Speculative to be sure, but I believe that if Jobs was alive we'd have a paper thin camera sensor because the bump would have been a nonstarter.

Same regarding your comment... I agree, the minimum QC does feel like it notched back a bit.

Agree and disagree on the "planning a trip" use case... as I'm sitting on a river cruise on an AI co-planned vacation (we found the cruise, AI set the daily itinerary).

Now the big (BIG) caveat is that I used Claude Code on my Max 20x plan from within VS Code. I have a fairly decent harness that I'd built and was sure to prompt it to run several subagents, including one that grounded walking times with Google Maps directly.

I'd say this is FAR beyond what the average person would do ("Hey Siri, plan me a trip to Prague") but also it shows that the models can do it with the right harness and guidelines. This wasn't that hard for me to do, so it seems to be more of a feature buildout ("the travel expert" AI) with a few markdown files than anything.

All told: web search for grounding times/locations, map grounding for walking paths and times, an adversarial agent to keep the model(s) honest, and a little bit of prompting and you've got a really great travel planner.

In short: the average person won't do this, but if I can build it in a few hours any of the 100% of people working at Apple/OpenAI/Anthropic who are smarter than me can build it and bake it into Siri (or ChatGPT, Claude, etc).

Or loop once/twice... it's a fun effect, but when endless I agree that it's distracting

I mentioned in another comment how hard it is for our brains to really comprehend the orders of magnitude difference between all animal cases (~680) and the former number of human cases (3.5M).

It would take ~5000 years at the current annual rate of animal cases to match the number of human cases just 40 years ago.

That's The Great Pyramid of Giza ago time... PLUS the amount of time since Michelangelo, Leonardo da Vinci, and Raphael roamed the earth.

The cool thing is that at a few hundred, one could theoretically* round up all (known) animal cases left. That's truly incredible work getting to this point if you think about it.

* Yes, geopolitical issues, geography, and plenty of other reasons might make this somewhat impossible... but the fact that we can actively picture a few hundred animals in our brains means that it's a very attainable goal.

Those are bonkers (low) numbers compared to the 3.5M (human?) cases if I'm to believe the GPs comment.

It's also crazy how much Mother Theresa's quote rings true, even in reverse ("If I look at the mass, I will never act. If I look at the one, I will.") When I initially read 3.5M cases, I thought "wow, that's a lot", and somehow the 445 animal cases in Cameroon felt (at first) more real and similarly "a lot".

No comment other than interesting how our human brains work and distort how numbers "feel".

Once my rational brain kicked in, realized that's over 5,000 years for the current number of animal cases to match the former number of human cases. The future is awesome.

My technical cofounder reminds me of this story on a weekly basis.

I love Anthropic's models but their realtime voice is absolutely terrible. Every time I use it there is at least once that I curse at it for interrupting me.

My main use case for OpenAI/ChatGPT at this point is realtime voice chats.

OpenAI has done a pretty great job w/ realtime (their realtime API is pretty fantastic out of the box... not perfect, but pretty fantastic and dead simple setup). I can have what feels like a legitimate conversation with AI and it's downright magical feeling.

That said, the output is created by OpenAI models so it's... not my favorite.

I sometimes use ChatGPT realtime to think through/work through a problem/idea, have it create a detailed summary, then upload that summary to Claude to let 4.5 Opus rewrite/audit and come up with a better final output.

> "LLMs are terrible at asking questions. They just make a bunch of assumptions and brute-force something based on those guesses."

Strongly disagree that they're terrible at asking questions.

They're terrible at asking questions unless you ask them to... at which point they ask good, sometimes fantastic questions.

All my major prompts now have some sort of "IMPORTANT: before you begin you must ask X clarifying questions. Ask them one at a time, then reevaluate the next question based on the response"

X is typically 2–5, which I find DRASTICALLY improves output.