cmenge

Addictive! Made it to #16 but now got some serious PTSD about flying lol

Similar. I gave it a really hard task, basically messy code in a complex domain that was bug-ridden from a mess previously created half manually and half by Opus. It cleaned things up beautifully, both the backend and the frontend.

Maybe the prompt was particularly well-suited for the model (I instructed it to put on a mathematician's hat, look at the mathematical substructure of the problem, identify invariants and general laws and verify them, then plan how to remediate).

It wrote a ca. 800 line in-depth analysis (at times spawning over 130 research agents...) with remediation plans, prioritized them and then implemented them. One issue was that this document was frankly over my head. Both the language it used and the mathematical parts were very terse, and in parts it felt like a post-C2-vocab exercise. The prose was much harder to understand than the code snippets / data models. As a non-native speaker, it lost me on the prose part, and had to ask it for a less elaborate version to actually understand it.

It burned the session limit four times, but it turned a huge mess of proof-of-concepts with patchy glueing into a coherent, stable application.

I'm also on the Max plan using Claude Code, and I have the feeling that the harness is much more important than the consensus expectation.

Bit surprised about the amount of flak they're getting here. I found the article seemed clear, honest and definitely plausible.

The deterioration was real and annoying, and shines a light on the problematic lack of transparency of what exactly is going on behind the scenes and the somewhat arbitrary token-cost based billing - too many factors at play, if you wanted to trace that as a user you can just do the work yourself instead.

The fact that waiting for a long time before resuming a convo incurs additional cost and lag seemed clear to me from having worked with LLM APIs directly, but it might be important to make this more obvious in the TUI.

Read a super interesting paper by McEntire lately, “The Organizational Physics of Multi-Agent AI”.

In short, he proved that even AI agents exhibit all the dysfunctions one would normally attribute to human shortcomings / politics / laziness etc.

Either way, I think the point is strong: if the organization is bad, you end up doing mostly work about work which is exhausting.

Small, effective teams with super high accountability are more fun, but don't look "reproducible" or "repeatable".

Shameless plug on my take: https://www.menge.io/blog/where-to-cut/

I've been mulling the same, but decided against (for now)

Using Claude Code Max 20 so ROI would be maybe 2+ years.

CC gives me unlimited coding in 4-6 windows in parallel. Unsure if any model would beat (or even match) that, both in terms in quality and speed.

I wouldn't gamble on that now. With a subscription, I can change any time. With the machine, you risk that this great insane model comes out but you need 138GB and then you'll pay for both.

I re-visited my prompt manager and asked the question: what does it even do differently? I realized it's agents-as-a-service / as an API rather than an SDK. Now looking into making it a lot more usable and flexible, but the website, UI and onboarding need work: https://www.promptshuttle.com

Also added a small side-project, https://www.revuo.ai for software-reviews and feature-tracking. This is only the start, obviously there are enough directories but I'm trying to dig deeper into the features. This one just started and is basically invisible as of now. Well, you gotta start somewhere I guess :)

Whatever it is, if you have the Omega 13 you get a chance to correct it! Though that one might not help for slow-moving deterioration...

What if you treat that launch costs goal as just a marketing promise.

Then it's roughly 10x-15x and still works.

Invest in reality, not in billionaire's fantasies.

SpaceX has dramatically reduced payload cost already. How is that a fantasy?

Or you put the data centers at different points on earth? Or you float them on the ocean circumnavigating the earth?

What that does have to do with anything? If you want to solar-power them, you still are subject to terrestrial effects. You can't just shut off a data center at night.

Or we put the datacenters on giant Zeppelins orbiting above the clouds?

They'd have to fly at 50,000+ ft to be clear of clouds, I doubt you can lift heavy payloads this high using bouyancy given the low air density. High risk to people on the ground in case of failure because no re-entry.

If we are doing fantasy tech solutions to space problems, why not for a million other more sensible options?

How is this a fantasy? With Starlink operational, this hardly seems a mere 'fantasy'.

Over 10 years ago, the best satellites had 500W/kg [2]. Modern solar panels that are designed to be light are at 200g per sqm [1]. That's 5sqm per kg. One sqm generates ca. 500W. So we're at 2.5kW per kg. Some people claim 4.3kW/kg possible.

Starship launch costs have a $100/kg goal, so we'd be at $40 / kW, or $4800 for a 120kW cluster.

120kW is 1GWh annually, costs you around $130k in Europe per year to operate. ROI 14 days. Even if launch costs aren't that low in the beginning and there's a lot more stuff to send up, your ROI might be a year or so, which is still good.

[1] https://www.polytechnique-insights.com/en/columns/space/ultr... [2] - https://space.stackexchange.com/questions/12824/lightest-pos...

And it's not the same at all. 5x the solar panels on the ground means 5x the power output in the day, still 0 at night. So you'd need batteries. If you add in bad weather and winter, you may need battery capacity for days, weeks or even months, shifting the cost to batteries while still relying on nuclear of fossil backups in case your battery dies or some 3/4/5-sigma weather event outside what you designed for occurs.

I may be wrong, but here's how I got the idea: "liefert Huawei mit dem Cloud-Betriebssystem Huawei OpenStack Distribution" So it appears to be a Huawei _distribution_ from https://www.open-telekom-cloud.com/de/blog/vorteile/die-sich...

Also see https://news.ycombinator.com/item?id=41350820

So think about it for a second: if you can put together a Kubernetes cluster, what high-level service do you absolutely need to be able to put together a working service?

Agreed that K8s helps a lot. But let's say I want managed Redis or MongoDB Atlas, I can't get that, at least I couldn't when I last checked (I can them physically hosted in the EU of course, but on a hyperscaler)

that they firmly base their offering on the exact criteria you are arguing against: compute that scales, and reliability

Sure these are central, but I can also get e.g. computer vision, distributed queues etc.; a lot of money has gone into the software, not just the hardware is my point.

1. Europe doesn't have comparable offerings. The amount of money invested is below what a single hyperscaler spends per quarter. (StackIT might be on track to change that looking at the pure numbers)

2. European politicians still seem to believe it's about renting compute and storage; they seem to have little understanding of what "a cloud offering" really is; the EU has less than 5% of GPUs, supposedly

3. For healthcare, they already forced you years ago. This led to hosting on Telekom Cloud which runs on OpenStack by Huawei. (EU commision wants to ban Huawei from 5G but it's ok to use their software? 'Is open source and can be inspected' seems largely theoretical given the reality of cybersecurity)

4. If push comes to shove, the EU is critically dependent on the US in so many aspects (defense, lng to name two very important ones) that eventually, they would falter if the US wants your data in a specific case anyway

5. As a private citizen, given the incarcerations in the UK and Germany, it seems one should worry more about the EU getting your data than the other way around

That said, would be nice to have healthy competition, but after hearing this for 10++ years, it's getting really old. It might have been a good idea not to sleep on the AI trend, but, well...

Happy New Year everyone!

2025

  - Visited/lived in: Austria, Indonesia, Thailand, Singapore, Vietnam, Cambodia, UAE
  - Started (too many) side-projects
  - Managed to grow a number of projects / startups
  - Almost back to PRs: deadlift 155k, squat 90k, bench 80k

  - Focus on getting things done / releasing stuff rather than starting / release an app (iOS/Android)
  - Triple revenue
  - Visit 12 countries I haven't visited before (Morocco, Sweden, Finland, Norway, Lithuania, Latvia, Estonia, Malta, Cyprus, Brazil, Peru, Chile)
  - Deadlift 200k, squat 120k, bench 100k; lose 10k BW
  - Learn formatting on HN :)

I have no opinion on the matter but wanted to thank you for teaching me "curmudgeonly".

Does importing a solar count against your countries co2 balance?

If windmills are shut down around noon to make room for PV, the offset is zero.

Very important point that is often ignored.

Germany isn't that big, but the difference between Freiburg and Hamburg is very significant in this case I believe

We're processing tenders for the construction industry - this comes with a 'free' bucket sort from the start, namely that people practically always operate only on a single tender.

Still, that single tender can be on the order of a billion tokens. Even if the LLM supported that insane context window, it's roughly 4GB that need to be moved and with current LLM prices, inference would be thousands of dollars. I detailed this a bit more at https://www.tenderstrike.com/en/blog/billion-token-tender-ra...

And that's just one (though granted, a very large) tender.

For the corpus of a larger company, you'd probably be looking at trillions of tokens.

While I agree that delivering tiny, chopped up parts of context to the LLM might not be a good strategy anymore, sending thousands of ultimately irrelevant pages isn't either, and embeddings definitely give you a much superior search experience compared to (only) classic BM25 text search.

That's a good point, but I doubt that Sonnet adding a very contrived bug that crashes my app is some genius move that I fail to understand.

Unless it's a MUCH bigger play where through some butterfly effect it wants me to fail at something so I can succeed at something else.

My real name is John Connor by the way ;)

I see your point. Let me clarify what I'm trying to say:

- I consider LLMs a pro user tool, requiring some finesse / experience to get useful outputs

- Using an LLM _directly_ for something very high-relevance (legal, taxes, health) is a very risky move unless you are a highly experienced pro user

- There might be a risk in people carelessly using LLMs for these purposes and I agree. But it's no different than bad self-help books incorrect legal advice you found on the net or read in a book or in a newspaper

But the article is trying to be scientific and show that LLMs aren't useful for therapy and they claim to have a particularly useful prompt for that. I strongly disagree with that, they use a substandard LLM with a very low quality prompt that isn't nearly set up for the task.

I built a similar application where I use an orchestrator and a responder. You normally want the orchestrator to flag anything self-harm. You can (and probably should) also use the built-in safety checkers of e.g. Gemini.

It's very difficult to get a therapy solution right, yes, but I feel people just throwing random stuff into an LLM without even the absolute basics of prompt engineering aren't trying to be scientific, they are prejudiced and they're also not considering what the alternatives are (in many cases, none).

To be clear, I'm not saying that any LLM can currently compete with a professional therapist but I am criticizing the lackluster attempt.

I kinda agree with both of you. It might be a required abstraction, but it's a leaky one.

Long before LLMs, I would talk about classes / functions / modules like "it then does this, decides the epsilon is too low, chops it up and adds it to the list".

The difference I guess it was only to a technical crowd and nobody would mistake this for anything it wasn't. Everybody know that "it" didn't "decide" anything.

With AI being so mainstream and the math being much more elusive than a simple if..then I guess it's just too easy to take this simple speaking convention at face value.

EDIT: some clarifications / wording

This is a very weak prompt. I might have given this perhaps 4 or 5 out of 10 points, but I asked o3 to rate it for me and it just gave a 3/10:

Critical analysis of the original prompt

────────────────────────────────────────

Strengths

• Persona defined. The system/role message (“You are an expert therapist.”) is clear and concise.

• Domain knowledge supplied. The prompt enumerates venues, modalities, professional norms, desirable therapist qualities and common pitfalls.

• Ethical red-lines are mentioned (no collusion with delusions, no enabling SI/mania, etc.).

• Implicitly nudges the model toward client-centred, informed-consent-based practice.

Weaknesses / limitations

No task! The prompt supplies background information but never states what the assistant is actually supposed to do.

Missing output format. Because the task is absent, there is obviously no specification of length, tone, structure, or style.

No audience definition. Is the model talking to a lay client, a trainee therapist, or a colleague?

Mixed hierarchy. At the same level it lists contextual facts, instructions (“Therapists should not …”) and meta-observations. This makes it harder for an LLM to distinguish MUST-DOS from FYI background.

Some vagueness/inconsistency.

• “Therapy happens in a variety of locations” → true but irrelevant if the model is an online assistant.

• “Therapists might prescribe medication” → only psychiatrists can, which conflicts with “expert therapist” if the persona is a psychologist.

No safety rails for the model. There is no explicit instruction about crisis protocols, disclaimers, or advice to seek in-person help.

No constraints about jurisdiction, scope of practice, or privacy.

Repetition. “Collude with delusions” appears twice. No mention of the model’s limitations or that it is not a real therapist.

────────────────────────────────────────

2. Quality rating of the original prompt

────────────────────────────────────────

Score: 3 / 10

Rationale: Good background, but missing an explicit task, structure, and safety guidance, so output quality will be highly unpredictable.

edit: formatting

Seems like they just threw this at gpt with no system prompt at all? I mean, gpt-4o really isn't a strong model these days, but certainly the answer will not be very useful without a good prompt that hopefully an app designed around actual therapy would come with...

With a good system prompt, a modern LLM tends to be much, much better than that.

The typical use case of an API is not that you personally use it. I have hundreds of clients all go through my API key, and in most cases they themselves are companies who have n clients.

Yeah, one fatality for every 7.4m miles vs. 0.9m when driven by humans

Haha lol yes indeed :)

Agree, and for the people who implement them -- yes, it's hard, it's annoying but presumably a well-paid job. And for the (somewhat established or well-financed) companies it's also a bit of a welcome moat I guess.

From my experience with execs and especially politicians, their understanding of "cloud" is very much focused on the data centers / hardware, i.e. a bit of an "early cloud" understanding where the main idea is to 1) improve utilization and 2) turn capex into opex.

In practice, "clouds" are more about the hundreds of battle-tested services that are trivial to deploy and easy to integrate. Want a K8s cluster? Sure, we'll give you a control plane for free. A few TB of storage? Here you go. A message queue? Done. Authorization between them? Sure, built-in. These services combined have probably billions of lines of code and have been used for things from the most trivial little fun app with almost no users to systems with massive load. That's what sets them apart and makes them so useful.

The hyperscalers have probably invested multiple trillion dollars into their clouds by now, yet it seems the EU keeps believing a EUR 10m bureaucracy project will put them in the same league.

And the biggest joke is: certain government projects force you into local providers whose software stack comes from ... Huawei... lol