StrauXX

6 ms·

31.
by StrauXX 5mo ago
LLMs don't tend to help much when solving challenges beyond their skill level. Either they one-shot a challenge, or thei are almost useless as a companion for them.
32.
by StrauXX 5mo ago
It is a hard requirement. Once you reach higher levels of challenges you spend most of your time reading through RFCs, web sepcs, Github issues, mailing lists, papers, random bugtrackers and library/framework code. There is no way to c
33.
by StrauXX 5mo ago
Which indications are that?
34.
by StrauXX 6mo ago
It is definetly cheaper now . What I want to say with this, is that token costs rising so dramatically that AI usage becomes uneconomical is not a high probability future. Even if AI subscriptions were sold heavily below cost (which is als
35.
by StrauXX 6mo ago
Self hosting at a reasonable scale is much cheaper than people think. I am running clusters of DGX Spark machines with BiFrost load balancers in our company and for client projects. They work flawlessly! 128 GB unified memory, Nvidia chip a
36.
by StrauXX 6mo ago
The party is called the "Christian Democratic Party" but in practice pushes no christian policies. 47% of germans are legally atheists. Only 5% regularly visit mass.
37.
by StrauXX 6mo ago
He has changed his opinion completely. Yes, the ratio has turned.
38.
by StrauXX 6mo ago
Only Stripe offers a service doing international VAT for you.
39.
by StrauXX 6mo ago
vLLM isn't suitable for people running LLMs side-by-side with regular applications on their PC. It is very good at hosting LLMs for production on dedicated servers. For the prod usecase ollama/llamacpp are practically useless (b
40.
by StrauXX 6mo ago
Northdata [1] does basically this, but mostly ingests European data currently. Perhaps they will expand to ingest US data as well at some point. Not affiliated with them in any way. I just use them to look into company structures every now
41.
by StrauXX 6mo ago
Do you plan on adding more models in the future? I would love to see how other OSS modles like Gemma, GPT-OSS and Qwen fare.
42.
by StrauXX 6mo ago
A lot of comments here are dismissing this post because the relevant code was isolated. But thats the exact same thing Anthropic did with Mythos! They describe their (very lean) harness in the Anthropic Red Mythos blog post. The harness fir
43.
by StrauXX 6mo ago
There are quite a few such startups already out there. Results are mixed so far. Though I believe they get much better over the coming months and years.
44.
by StrauXX 6mo ago
Switzerland is one of the least homogenous countries in Europe! Four languages, relatively weak federal government with strong local (canton) governments. Most state services are handled on a localized (canton) level.
45.
by StrauXX 7mo ago
I had the same impression. The introduction reads really unprofessionally.
46.
AI Wafer Timeline(xcancel.com)
2 points by StrauXX 7mo ago|0 comments
47.
Organizational Politics and the Security Program(philvenables.com)
3 points by StrauXX 7mo ago|0 comments
48.
by StrauXX 7mo ago
If your requirements are just to load balance between selhosted AI servers: nginx. If you want a more thorough system with configurability, logging, etc.: Bifrost from MaximAI
49.
by StrauXX 7mo ago
LiteLLM isn't a good choice for a proxy in any case. It introduces a lot of lag amd latency and the features are often half baked. To me, it looks like a vibecoded application without a product owner. And the code itself isn't ver
50.
by StrauXX 7mo ago
You shouldn't! The official writings explicitly discourage creating third and fourth layer diagrams!
51.
Clearing Member 991(no01.substack.com)
1 points by StrauXX 7mo ago|0 comments
52.
by StrauXX 7mo ago
The "market" isn't a zero-sum-game though. Polymarket is.
53.
by StrauXX 8mo ago
They are! This Tesla feature is illegal to use in Germany and Austria. [1] (in German) https://www.derstandard.at/story/3000000215526/aktueller-fal...
54.
by StrauXX 8mo ago
This tracks with the way a lot of heavily vibecoded projects have issues with beeing feature heavy, while those features often don't fully work and most importantly don't fit together cohesively. In other words, the quality is low
55.
by StrauXX 8mo ago
That's "just" the interpretation of TU Berlin. Here is the FOI request that was never answered: https://fragdenstaat.de/anfrage/onlyoffice-sanktionen/ Just to be clear: TU Berlin is probably right.
56.
by StrauXX 8mo ago
OnlyOffice had some controversy around being owned and operated by a Russian company through shell companies. They might even fall under EU sanctions. There is an open German information request to the government that was never answered.
57.
by StrauXX 8mo ago
Do you have a reference with more information on that?
58.
by StrauXX 9mo ago
Simple. You multiple instances with different flags covering different threat models. RCE, file read, etc. You then expose a webapplication for every instance that lets users control only those curl flags, that must be safe to be user contr
59.
by StrauXX 9mo ago
The process of finding bugs is still slow and time consuming. The kinds of vulnerabilities you find in codebases like cURL are still beyond AI. Binary exploitation is still a human only field.
60.
by StrauXX 9mo ago
The solution for this, IMO, is flags. Just like with CTFs, host an instance of your software with a flag that can only be retrieved after a successful exploit. If someone submits the flag to you, there is no argueing about wether or not the

More ›