antupis

I think it's more that agents gaslight you to believing that timezone from Baker Island and Howland Island is most important issue ever and needs that 1k exception.

Thing is that it does it silently, I only noticed this because randomly opencode summaries were something like sorry I can’t summarise this.

Issue with Opus is that you can’t do regular pr reviews where you say please try find security issues at best there is silent refusal and worst you get downgrade.

“ You typically spent years building before you had a platform, then lack the enthusiasm from senior leadership to invest the next 12-18 months with no new features and hard to quantify - maybe no - returns.” —> this is failure from the team and especially people who should do product management job on that team.

And I feel that stuff like agile is now even more important than before. Now it’s role should be more about protecting customers from bad changes and following whole processes through where you actually measure the impact. It’s just too easy to do those 100k lines changes to products because it’s some vice presidents pet project.

At least in my app I have Argon, but that said you can pretty easily prompt it if you have gcp subscription or through openrouter.

There is difference between blocking all content and just making model non-savant in some area.

They are also supply cuts eg Russia is pretty much screwed and I bet that they don't return ever level that they were producing oil + natgas in 2021. Iran and UEA has lots problems because military strikes but those probably return normal when crisis is over.

Well they dull emotions as you probably don't snap and hit some bystander/teacher/parent/friend, but also you probably feel as normal people feel as 'fine' which might be issue for some people eg if you are musician and you need strong emotions to perform. But good thing with stimulants is that effects are pretty short term and if you feel that they dull emotions too much for now you just can stop taking those for moment and then continue when you need it.

Well cutting is happening right now and very large scale. USA is kinda laggard here but eg new electric car sales is up 70% in Europe or new Solar installations are up 50% in Africa.

I think there running both will be winning combination scripts/classical ML do large scale crunching and when they find some anomaly or some threshold creates alert. Agents will create initial report and maybe create first pr for developer.

Can you share that vim plugin? Now running opencode at Ghostty and looking diffs at github and rarely when need to do code changes manually I am using vscode but proper vim plugin would be godsend.

Or humans just prompt wrong thing or people actually don’t like feature.

For my own benchmarks copilot cli uses 1.5x more tokens than opencode.

Thing is that you cannot do 24/7 oncall with two person and Opus 5.5 . Also bus factor start to be pretty big if there is only two persons etc. There is still reasons why FAANGS are so big.

Yup, and if someone is looking for a non-AI startup idea, take a use case with lots of data and a read-heavy workload, and do pretty much what Turbopuffer did for text and vector data. That seems like a very good way to go.

But issues is that for the valuations you need models that work without human expert and Opus 5.5 isn’t there even easy stuff like programming.

Pretty much this, I don't just see how ASI could happen with current LLMs without big theoretical breakthrough so that's why 'phasing the frontier' and 'distillation attacks'.

Flash thinks much more so it’s pretty much line with Sol for performance. That said I like flash coding style much more than OpenAi models.

I think it's more about money and scaling, bus factor is pretty big if run very lean organization eg whatsapp 2014 and that point all those developers are kinda your cofounders and probably start asking much bigger piece of pie. With small teams you kinda trade scaling and availability to velocity. It's much easier to ship but running oncall 24/7 with small team is just nightmare.

I think we are starting be on that territory that regular software development is suffering, current models are great for benchmarks and one-shots but in daily development models are too eager and try to force patterns like excessive tests in every turn.

I think automation is coming but it will be way more gnarly than frontier labs want public to believe. Value is just too big, when you can automate most of eg customer support it will create huge savings and same time customer satisfaction will get better.

we aren't anywhere near lights off software factories, and writing software is easiest domain for modern llms, you can verify results rather easily, plenty of training data etc.

I think it’s more that pushing frontier is extremely costly and there is no free lunches in same way as 2024.

My main issue with rtk is that rtk somehow messes modification and agent start polling same tool continuously.

Sweden main issue is that there is no transmission capacity between North and South, like now prices are 2x at south.

Yup and this propagates those clumsy if this_new_code_branch: actual_code_that_matters else: old_legacy_code_that_should_not_be_there

Yup, new SOTA models especially with high/xhigh/max reasoning too often overengineer solutions, good for benchmarks that usually measure task completion, bad for normal development where you don't want 'rewrite in rust and 1k LOC unit tests' style solutions when agent does mundane bug fixes.

Mistral just needs to good enough category think all those flash models or Qwen3.8 27b which they sadly aren't at the moment, that plus being European lab will mean that they will have very nice business. Even now these SOTA models feel too overkill for most tasks.

Yup like you go far enought you will pick lots of transferable skills like data cleaning in the case of LLM, DOM parsing case of web browser or welding case of rocket.