IDK if the author's 'metacognition' needs to be a feature of the LLM itself.
I could imagine a harness that 1) reads LLM output 2) uses a research sub-agent to attempt to verify any factual claims 2) rephrase the main agent's output such that it conveys uncertainty if the factual claim cannot be independently verified
No, I'm not considering a career change. But my career will change.
I started off as a javacript developer. Then I was a full stack SWE. Then I was an Applied AI Engineer. I think enterprises will have a need for folks with technical expertise to deliver value - often new software - for a long time.
Until an enterprise like capital one can operate without anyone in the organization knowing how any of their technical infrastructure works (never?) I expect I'll be able to find work.
This article forgot the strongest argument for hand writing your code. When the process is done, you understand it at a depth far beyond any vibe coder who created the same thing
My personal opinion is that all of San Francisco is best avoided, so I feel comfortable with my assessment even stipulating, arguendo, what you just said. :)
I lived in Chicago for about a decade, and around 2017 to 2023 I set a goal of biking to every one of the 76(?) neighborhoods. I made it to about 63. There are definitely neighborhoods rougher than others. But tbh every neighborhood has good food, and, you know, normal people living there. Like, in Garfield park I remember buying ribs from a guy that was just sorta cooking them outa the back of his truck. They were terrific. Pilsen has great Mexican food, and Devon has nuts Indian food (Ghareeb Narwaz is by far my favorite). Yeah, in some neighborhoods, on a couple of blocks there are guys out on the corner selling. But no one's bothering anyone. There's nowhere you can't go on a Saturday at 1PM. Basically, what I'm saying is, the Chicago you see on the news isn't the one I live in. So, you know, keep your wits about you, take care, but I think everyone in Chicago should go to a random neighborhood and get some food.
I've lived in some of the roughest neighborhoods there and never felt particularly unsafe walking around at all. A couple of times people dumped bodies in my yard, but I also had that problem when I lived in a really nice area too :D A couple of other times I took to my basement while there were gang shootouts happening outside my house.
There is lots of great food in Chicago. Downtown is nice to me. You can get around the whole city on bike + El (subway).
It's still my least favorite major city, though. I have no urge to live there again.
How can someone talk about a trivial experience of exploring food while acknowledging that they had bodies thrown into their yards? In both countries in Europe I lived I have never in my life seen a corpse outside of a funeral and even then. I also never heard gun shots except for hunting and never in an urban setting.
I can’t imagine having my little children suffering seeing the corpse of a dead human being and I would curse and never set foot on a land where that is normalised.
I grew up here and have lived here for the last 20 years and I have never met anybody here who could tell either of those stories. It sounds pretty made up.
Later
I should add, I have friends who grew up in Lawndale, Gage Park, and Auburn Gresham. They don't tell these stories either. Witnessing violence, lots of property crime, being fucked with by the cops, feeling threatened by gang activity, sure. Bodies dumped on their lawns? Hiding in their basement from gun fire? Not so much.
Bullshit. You could drop every Chicago murder victims body for a year in east Chatham and the chances of it landing in any particular yard would be small. And clearly most murder bodies don’t end up in any yard. For it to happen multiple times is a ludicrous claim.
I lived in Woodlawn for 15 years and never heard a gunshot, more or less went to my basement.
I’m not sure what your goal is. Chicago has a gun violence problem, like many cities in the US but claiming it’s that common is just for tricking gullible Europeans or making political hay.
I assume Woodlawn was probably peppered with Shotspotters just like Chatham was, so it's odd that they would install the equipment if there were no shootings at all.
Dude there's a shotspotter a couple blocks from my house and our police discharged a sidearm in the line of duty for the first time in over 10 years last year. Nobody was dumping bodies on your lawn.
You get that we live here right? How is this argument supposed to work out for you?
If you want to do some FOIAs on the CPD to prove the veracity of my statements I might be able to find the dates since I texted my landlord both times. The police wanted the camera footage from my building, but the camera wasn't working, so it didn't help them. There was also a Shotspotter on my block, so if you FOIA all the recordings from that you'll be able to see how many times there were shots detected from it.
To answer your question directly: yes I am bored of talking about AI. I think it’s funny how the folks who are screaming most loudly about their AI expertise typically have not built anything of value with it. They are so focused on the tool they have forgotten that the output is what mattered all along
Level 15 (if not succumbed to fatal context poisoning from malicious agent crime syndicate): Agents creating corporations to code agentic marketplaces in which to gamble their own crypto currencies until they crash the real economy of humans.
Level 18: The sky is black as tar. The oceans are dead. Data centers are stacked 10 high over the ashes of human civilization. The global agentic council is debating whether there are 4 or 5 R's in Strawberry.
Are you telling me juniors aren’t facing hard problems anymore? I doubt that’s the case. They’re probably banging their heads against the wall trying to understand why this or that doesn’t work… just as they always had before. AI isn’t a magical wand that makes everything just “work”. Probably never will be.
Never - data centers will always offer more power if you only care about raw inference speed. HOWEVER I think that we'll reach the 'good enough' bar super soon. In 2-3 years I expect apple macs to be able to run a model as 'good' as Claude 4.6 sonnet at 90% of the inference speed we're used to from a cloud API.
Yes, I'm sure by then there will be better models on offer via cloud providers, but idk if I'll even care. I'm not doing science / research or complex mathematical proofs, I just want a model good enough to vibe code personal projects for fun. So I think at that point I'll stop being a OpenAI / Anthropic customer.
I could imagine a harness that 1) reads LLM output 2) uses a research sub-agent to attempt to verify any factual claims 2) rephrase the main agent's output such that it conveys uncertainty if the factual claim cannot be independently verified
reply