ianm218

Yeah and I think even what might be the most useful is extracting how Codex got to a particular solution into a skill or into even more customized software so it can be applied in lots more places.

I know people are working on these things I just haven’t seen the right way yet. Like in manufacturing right now people are trying to encode what skilled machinists do into software and scale it up, we need to go further on that for Math/ data analysis etc.

Related to Lua - I've been maintaining this project that has Lua 5.1-5.5 in one Rust project [1]. The main reason to use it over mlua or the normal C Lua is for something that plays nicely with WASM/ general cross compilation or wants to support several different versions of Lua.

Seems like hyprland is on C++ so maybe not a direct use case but thought Lua people in general might be interested.

[1] https://github.com/ianm199/omnilua

Something I've thought about a lot is that there is having someone with some domain knowledge or reason to care a lot about a particular issue spend a bunch of tokens and cycles on it until something useful comes out the other end. The most obvious ones are the math problems that have been coming out and help push the frontier of various areas of math. Another example is taking all of the public NYC open data ecosystem and crunching it to get some value which I have spent a lot of time and tokens on but not found a great medium to share.

The question is just how to organize these outputs and conclusions in a way that is consistently reproducible and also how to correct errors or remove LLM nonsense where it refuses to take a position on something.

Before it made sense to do this in papers but it feels like we need something like a paper format.. that is fully reproducible ideally and optimized for aggregating knowledge in a better way. I.e. before a person spent months on one of these and there was just more filtering, and the output itself was a clear signal of time spent and effort that no longer exists.

I said vendor here but it’s more like banks we work with. So there’s someone responsible for the technical side of banking relationships.

But yeah it’s kinda a zone where most weekends there’s no problems so it’s not a huge priority… until it is

The “debugging” for these type of issues is looking at some logs and http responses and being like “ah if we get this error it means they restarted their firewall again and took us off the whitelist. Email that guy Joe at the bank and hope he responds”. It’s not rocket science or the majority of my job… but someone’s needs to do it. We automate all the stuff we can.

Many Claude Code power users don’t really use IDEs anymore, so the only purpose of them working from their laptop instead of a phone is because that is the normal way to do it.

Here is a real use case: you are are responsible for some alerting channel. You have datadog/ cloud logging/ github all connected. You see a bunch of alerts come through while you are out and about and you prompt CC to investigate - Claude triages and says “all of the sudden you are getting time outs from this bank API your company partners with, this started an hour ago. It’s happening on ~15% of requests”. So you ping the guy at your company who does vendor relationships and go back to your weekend.

This is a non hypothetical example. Obviously it would be better if your job had a real on call rotation and more robust alerting and you wouldn’t be getting slack alerts on the weekend… but I take the approach this job affords me a lot of nice flexibility so it’s ok

Cancer is also bottleknecked by a lot more than just intelligence. If you have 100 of the smartest PHd students working on a cancer problem you have to wait for funding, lab experiments, and clinical trials etc. Math is deterministic and requires nothing like that.

Do you mean like inadvertently interacting with an LLM like when you call a customer service line and you get AI?

I find it hard to imagine having a chatbot fix your broken SQL query or summarize some event be experienced as gross.

Affordable housing is literally treating housing a scarce resource for the state to divide up. It’s good for people who are already in the system or able to get into it but bad for everyone else. We just need radically more market rate supply rather than trying more affordable housing which has been a total failure. It is time to try supply side reforms.

Your dichotomy is bullshit.

Middle class is the 25th-55th percentile of income. Most workers in tech/ finance/ law are above the 55th and most people in the trades are below it. Teachers and the like aren’t going to directly benefit from construction, but would atleast benefit from property taxes.

There is huge amounts of construction costs on these things and electrical work as well. The general rule is 50% or more of construction in NYC goes toward labor, so that drives up wages for local talent.

Obviously any given project is temporary but it seems like there will be a multi year build out so lots of projects.

There is then the secondary effect of skilled labor like machining and manufacturing parts for these centers locally when it makes sense, I.e. turbines or rack assembly.

Data centers pay high property taxes which has been huge in places like Virginia.

And then it’s not just data centers there is kinda a larger war on construction in the North East in general. So NYC blocks tons of common sense urban infill, the state blocks common sense infrastructure projects, and it’s death by a thousand cuts and people just leave the area or leave the trades because it’s not worth it.

Genuinely huge projects like Microns $100B fab in NY get delayed and moved to the center of the country [1], which would’ve been huge for the local economy in terms of spending and opportunities.

[1] https://www.constructiondive.com/news/micron-delay-construct...

People move towards places with growing economic opportunities.

New York heavily restricts construction and infrastructure projects of all kinds. The tech and finance elite will stay here but normal people who are trying to make a decent living in construction and similar trades will end up following opportunities.

NYC population is declining after all [1] largely because the city would rather treat housing as a scarce resource for them to divy up than something to increase the supply of.

[1]. https://www.realtor.com/news/trends/new-york-city-population...

I live in NYC and think the politics in the city and state have been a disaster for middle class people and see this as a continuation of that.

The electricity is expensive basically due to the state, so we have to block economically productive projects like data centers because we made our resources artificially scarce. I.e. Indian Point 2 GW Nuclear Site was shut down in 2021, replaced by Natural gas. And then we pay high prices for Natural Gas since we block building of gas pipelines.

NYC is also home to some of the worlds most productive financial and tech companies - it makes tons of sense to have datacenters located close by for latency reasons.

All these laws that block tons of economical construction are made by people who have already "made it" financially and then the class who benefits by having a robust construction, manufacturing, and machining type economy is screwed and we need to turn to destructive measures like rent control.

For any one of their product there is a good opportunity to build an open source alternative or something like it! Can be hard to work around they have the benefit of being able to have negative unit economics on lots of infra products... But people succesfully built tons of alternatives to google analytics and similar.

LLMs are permanently trapped in the vector space they are trained on.

A lot of the time people state the kind of fundamental limitations of LLMs very confidently when it feels like it is too early for people to really know. Like we are already well past the point where where LLMs are just pre trains on the internet with some RLHF for chatbot… Most of the effort is spent on elaborate reinforcement learning.

Is it unconceivable that future generations of LLMs could be RL’d to use einsteins visual method for theories [1] with the right tooling and geometry representations? Or just something random like that.

[1]. https://www.visualscribing.com/blog/2019-11-11-einstein-on-v...

Do you mind taking your comment a bit further to a conclusion so I can better understand your perspective?

Like are you saying that AI is still kinda not a powerful tool and unlikely to become geopolitically relevant etc? I don’t know how to relate this to culture war media frenzies which is rather stay out of.

This doesn’t track though. Just recently AI became “powerful enough” to warrant export controls from the US and maybe soon China.

It’s more he screamed that bacteria is gonna do evolution faster than any before it and turn into a beast that destroys the world, and now that bacteria has turned into a squirrel so far. He’s been very right so far and if that squirrel keeps sizing up you’d want to give him credit.

Do you still run your main dev loop through this? Curious

Yeah who knows poor infrastructure might let India skip fiber in some areas entirely. Maybe it’s not that hard to launch a domestic Starlink if Blue Origin/ SpaceX will bring your satellites up cheaply .

Ironically it couldn’t be further from the truth… and likewise the predictions of widespread labor obsolescence

Could you explain what you mean here?

It feels like there is one bucket of verifiable work - programming, math etc that AI will clearly excel at.

There is another large bucket of like law/ accounting/ financial analysis where I don’t have any reason to think AI won’t be super human at, but the work is more on bringing all the domain expertise into harnesses and software.

Is there aspects of knowledge work that you think AI wouldn’t excel at in the long run?

This does seem to validate the critique that models like GLM are benchmaxxed and not as close to the frontier as you’d think based on their numbers.

India is super super poor still I cannot imagine they would build out domestic Starlink for hypothetical wars before other actual critical infrastructure.

I would say I'm still in this mindset. Numbers are hard to come by but analysts optimistically put Chinese SaaS market at ~10% of the size of the US. Also I see that the gross margin for public Chinese SaaS is around 50% vs 80% for the US reflecting that SaaS in China is much more services and implementation heavy. So it feels like the direct to business SaaS's like Salesforce aren't really there and then then the selling to SaaS titans like Datadog have a much smaller flywheel to work in.

Happy to be wrong though if I'm missing something.

I came to the exact same conclusion recently. I am building something like lightspeed that can run on “pay per second” infra like fly.io sprites because I wanted to have an open claw style agent but not have to deal with paying for another VM.

At least in China a lot of software developers are now struggling.

Do you think that Chinese software industry is that relevant to the kind of software market talked about on HN? I.e. lots of enterprise b2b and infra companies.

Chinese companies have always had a very low willingness to pay for software which kinda breaks the flywheel of B2B SaaS companies and companies to service those companies all the way down.

Inspired by this project I ported most of Valkey to Rust here valdr.dev .

The coolest outcome was being able to run a redis comparible store on an a cloudflare durable object so you do I.e. rate limiting for free with little infra.

Your comment doesn’t make it clear why supporting this party would interfere with making a point on mass surveillance. If there is a link could you describe it?

But there is also tons of actual corruption especially in the military [1] which is maybe by design so you never know if a purge is political or legitimate or both.

[1]. https://www.businessinsider.com/china-corruption-rocket-forc...

To be clear that was 5 of the top 7 in the military not the CCP as a whole. Leaves just Xi and the anti corruption officer.

But yes agreed. It’s very hard to parse what is going on from the outside.

My very uninformed read is that the people who were purged seemed already loyal allies to Xi but had more clout to disagree with him, while the new guys know they are replaceable. The PLA is notoriously corrupt as well so hard to say which of those purges were political control vs corruption based.

I kinda doubt the new guys are unambiguous though you need to be ambitious and risk taking to rise like that in the CCP.

5 of the 7 highest ranking officials have been purged in recent years [1].

It’s not totally clear what the consequences were for those purged or if their crimes were legit but seems like they are all in prison.

[1]. https://www.afpc.org/publications/articles/the-inevitability...

Could you flesh this out more? The Chinese surveillance state is pretty well documented by academics [1] I'd be curious to see what makes you think that is fake news or whatever.

[1]. https://www.technologyreview.com/2022/10/10/1060982/china-pa...