Walk outside so that the morning sun can directly hit your eyeballs. Don't stare directly into the sun. Try to look at something a little bit off axis. Your eyes should feel somewhat uncomfortable if you are doing this correctly. Do this every day for 3-5 minutes. You can accomplish this incidentally by just doing normal stuff like taking care of your own lawn and going for a run.
Light coming in from a window is generally filtered. Light being bright can be very deceptive. Even if the temperature is correct. Your eyes crave that deep infrared energy that LEDs and energy efficient windows actively reject.
> I tried the cheapest provider on openrouter and burned through $50 in a few days.
I wish I could observe how some of us are using these tools.
I still struggle to spend $50 in tokens per month, and I exclusively use prepaid API tokens. This is in support of personal projects and two clients. There are billing cycles where I might spend upward of $400, but this is maybe once a year. This is offset by months like August wherein I spent $12 in tokens.
The other advantage with prepaid is that it handles the other direction much better. I don't even know what a quota limit feels like. Being blocked for hours is way more expensive to me and my clients than even $500/m. Losing an entire business day over this wouldn't work out.
I've managed to convince some others to try the same thing. $200/m flat fee is a pretty extreme constant expense if you can be more clever on average.
I think a lot of people are getting pushed around by FOMO effects into spending money on pointless subsidized tokens and have (valid) fears that if they don't maintain the same apparent economic leverage as their peers that they will be left behind. This isn't actually the case, much like lines of code are a really poor indicator for the quality or productivity over a codebase.
I think it depends on how many threads you have running at the same time. I have Claude writing a compiler in one window, a ui framework for the language in another, an application using the installed versions of compiler and frameworks in another, and a ui designer (an Interface Builder lookalike) in another keeping up with the framework.
Each of these has a file it listens to in ~/tmp/.io and whenever one needs something from the other, they message each other via that file. Tasks can bounce back and forth as issues are resolved and tested. At the same time, I keep each busy with a list of tasks
I can fairly easily run out of my $200/month subs every week, if I let Fable be the default model. With Opus it’s less likely. If and when I do, I just have an alias ‘claude.ds’ which fires up Deepseek instead, and burns through far less money, though I don’t think it’s as good at solving problems, just MHO.
> whenever one needs something from the other, they message each other via that file. Tasks can bounce back and forth as issues are resolved and tested.
Why isn't this just one coherent agent loop with subtools/agents as appropriate? If these tasks are related in some way, having a single context would probably make it go much better.
The freewheeling messaging part is where the token bloat is coming from. I suspect that for some of us this is actually the point. I think it's a mostly form of entertainment to do things this way. The next logical step from Factorio gameplay.
Parallel agents remind me a lot about multi core compute. It's incredibly easy to take a single core product and make it run much worse across a lot of cores.
Which models do you primarily use, and can you very roughly list your process? Agentic coding in VSCode with tons of MCPs or... something else? Do you include lots of images or have large codebases? Which agentic harness are you using?
I also find that it's easy to spend like that, but also easy not to with little impact on productivity. At the current moment I'm stuck with rider + copilot (not ideal), but e.g. using GPT 6.1 luna is really, really cheap, and lots of tasks are quickly and decently dealt with even at lower reasoning levels, (added bonus of having low latency). And that model is so cheap, I can't see a hitting 1500$ at api prices realistically - not even close. But it also depends on the harness and codebase.
I use opus primarily, on a mix of pure coding tasks, and log parsing / incident investigation.
I don't have the mental capacity to do a lot of context switching between active work streams, so I'm not doing stuff like leaving a big agent workflow running while doing other things.
> If you've ever made the jump from PC to Mac (or vice versa) at work or home
I make this "jump" about 20 times per day. I have iOS, MacOS and 2 windows machines at the moment. Very frequently I will RDP into the windows machines from the MacBook.
> I don't think the challenge with speech to text was size of the binary
It's a very compelling aspect of the problem.
If you can get a model under certain size thresholds, that means you can eliminate latency domains. For example, if the model is able to fit entirely inside L3, the latency of servicing requests drops by an order of magnitude (or better) compared with a model that resides in L3+DRAM.
It's true that L3 is an order of magnitude lower latency than DRAM, but that's mostly going to translate into a throughput difference rather than a latency difference, when looking at the system as a whole.
If we take some time to understand how HBM memory is manufactured (with particular focus on yield risk for final packaging steps), we will hopefully learn that the current capacity crisis is not bullshit.
I guarantee Micron & friends are not intentionally orchestrating their business such that they would suffer a massively reduced chance of yielding on a per-die basis. Unless someone is actually buying HBM devices, they are not going to be making them. These are not a commodity that can be speculatively manufactured in any economically rational way.
The biggest issue with batteries is that they are not a prime source.
In a healthy grid with a diverse generation mix, batteries are almost certainly the perfect option for peak demand handling. The problem is that many grids are not healthy or do not have very diverse generation mixes.
You are effectively borrowing from yesterday (or last hour) to pay for today with batteries. Which is amazing, until it isn't. The Texas winter crisis comes to mind as an example where batteries would be regarded with intense derision. I think the current capacity market accreditation process is massively underrepresenting the tail risk of a black swan event. Four hours of battery storage should not be in the same room as a gas turbine when we are talking about capacity and multi-day emergencies.
Four-hour grid-scale batteries wouldn't have helped the Texas winter crisis much. But something like a 10 or 20kwh home battery would have helped a lot (it wouldn't sustain normal energy usage for that long, but the individual home owner could choose to ration the available power for critical functions).
Making demand responsive to supply costs is critical for increased efficiency. Part of that is making appliances smarter and educating users on loading washers/dryers/dishwashers and letting the machine handle energy usage. There are efficiencies in forecasting demand so supply can be negotiated and scheduled, and also being responsive to short term load shedding requirements.
It looks to me like it might not be worth worrying about this anymore. If you've ever owned a plasma TV you would be rolling your eyes at these comparisons.
My 2007 Pioneer Kuro Elite 50" plasma still displays a perfect, deep, gorgeous picture that sometimes looks to be almost 3D, so much so that I won't replace it with a new 4K TV until it finally dies. If it ever does....
> idealistic open-source maintainers build basically for free while scrambling to even make basic income.
Maybe these maintainers could simply cease doing all this free work? There are other ways to make a living. Perhaps even doing the same thing but with better negotiation skills. No one is forcing anyone to do this. The reality is that if 100% of the tzdb and SQLite teams disappeared today, Microsoft and Google would have these projects back under stable operation before the weekend.
We cannot expect a realistic economy to arbitrarily compensate people who originally did not seek compensation without a lot of really bad side effects. Does anyone on HN have any actual evidence that the maintainers of truly load bearing infrastructure (billions of devices+) are genuinely struggling economically?
There are a lot of degrees of "off grid". Society also includes less dense (non-urban) regions. Being miles away from your neighbors permanently isn't a crazy idea if you live in Kansas, yet they still have a coherent civilization.
It isn't very hard to reinterpret this critique as a kind of crab bucket mentality. God forbid someone goes off on a crazy boat trip for a few months without first consulting with "society".
Light coming in from a window is generally filtered. Light being bright can be very deceptive. Even if the temperature is correct. Your eyes crave that deep infrared energy that LEDs and energy efficient windows actively reject.
reply