Hacker News: jascha_eng

New comment by jascha_eng in "Claude Code: Anatomy of a Misfeature"

jascha_eng — Fri, 17 Jul 2026 17:06:07 +0000

Until the classifier is wrong or also prompt injected. the classifier is just as vulnerable as the model itself is. Yes it is harder to break but trying to make a nondeterministic tool deterministic by adding another nondeterministic one on top just reduces the chance of something going wrong.

Tbf as long as that chance is low enough it doesn't matter in practice, but I have definitely seen the classifier approve things that were questionable, and I've also seen it decline things that were obviously okay.

New comment by jascha_eng in "OnePlus halts operations in USA and Europe"

jascha_eng — Thu, 16 Jul 2026 13:52:30 +0000

But a pixel is quite a bit more expensive no? At that point you can consider an iPhone?

New comment by jascha_eng in "OnePlus halts operations in USA and Europe"

jascha_eng — Thu, 16 Jul 2026 13:18:23 +0000

Sad I have a 6 year old oneplus and was looking for a new phone somewhat soon, would've considered them again for sure. Any alternatives? They always had a reputation for me for being a great no fuss, little bloat and simply fast android phone.

New comment by jascha_eng in "The real prices of frontier models"

jascha_eng — Mon, 13 Jul 2026 21:43:09 +0000

I mean it might lead to better performance on the model side. So the tokenizer is better but more expensive.

New comment by jascha_eng in "The real prices of frontier models"

jascha_eng — Mon, 13 Jul 2026 21:40:43 +0000

Aside of the claudeisms and the obvious AI smell, it overexplains everything and doesn't come to any useful conclusions. It's just not a good post.

The nudge to think about both "tokenization as variable" as well as actual tokens consumed per task is still good.

New comment by jascha_eng in "Ask HN: What Are You Working On? (July 2026)"

jascha_eng — Mon, 13 Jul 2026 02:25:22 +0000

Slowly improving the UX on my SQL review/approval tool: https://github.com/kviklet/kviklet

Also finally closed the first real customer on it recently!

I want to get through a large chunk of the open issues the next few weeks and then spend some time building agentic capabilities for it. I believe a central place to configure database access for your dev team without having to share passwords and with sensible review policies should also help e.g. if claude needs to access production data to validate a premise.

Still have to figure out the right UX though not sure the agent should have the exact same review requirements that a human does. Maybe it needs to be configurable separately

New comment by jascha_eng in "GPT-5.6 Sol Ultra will be in Codex"

jascha_eng — Tue, 07 Jul 2026 18:00:11 +0000

im talking about anthropics pricing

New comment by jascha_eng in "GPT-5.6 Sol Ultra will be in Codex"

jascha_eng — Mon, 06 Jul 2026 11:05:51 +0000

Apparently not? At least our it guy said after 150 people or so you have to pay for enterprise which is pay per token for everyone.

New comment by jascha_eng in "GPT-5.6 Sol Ultra will be in Codex"

jascha_eng — Mon, 06 Jul 2026 06:59:20 +0000

Not including their best model in a max subscription would otherwise be truly a good reason for once to consider going back to openai for me. I'll at least try it.

New comment by jascha_eng in "Leaking YouTube creators' private videos"

jascha_eng — Sat, 04 Jul 2026 17:58:02 +0000

It's great software in the sense that it makes a shit ton of money though. In the end software that doesn't get used and doesn't make any money but has no bugs is not valuable either.

Not saying that this is the trade off you have to make but if you have a working mode in place that achieves usage and money somewhat consistently i can understand being hesitant about changing it to optimize for less bugs instead.

New comment by jascha_eng in "Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers"

jascha_eng — Thu, 02 Jul 2026 06:42:00 +0000

I mean these were all solved before I assume so 100% not the same human ofc but models are expected to be good at a variety of code bases while human can specialize in one and learn. I think it's fair to compare to an individual that is used to working on a product.

I'm more interested in how fable would do

New comment by jascha_eng in "Fable 5 will default to Opus 4.8 for coding tasks"

jascha_eng — Wed, 01 Jul 2026 19:35:46 +0000

Can't wait for China to pull ahead so we have an end to this bullshit

New comment by jascha_eng in "GLM 5.2 beats Claude in our benchmarks"

jascha_eng — Mon, 29 Jun 2026 06:33:39 +0000

Shorter sessions more often doing a /clear etc. save a shit ton of tokens. I pay 100 bucks a month but barely use 30% of it most weeks.

New comment by jascha_eng in "Founding a company in Germany: €9600, 152 days and I still can't send an invoice"

jascha_eng — Thu, 25 Jun 2026 18:12:55 +0000

I mean if you have an idea or project that convinces you its worth to found a company, a thousand bucks is not a great hinderance. If the 1000 bucks are a hinderance you probably shouldnt found a company. It's a decent filter if you are serious about it.

New comment by jascha_eng in "Founding a company in Germany: €9600, 152 days and I still can't send an invoice"

jascha_eng — Wed, 24 Jun 2026 13:10:24 +0000

idk I founded a UG within a few weeks recently. Yeh it costs me maybe a thousand or so in notary fees but that's it.

New comment by jascha_eng in "Project Fetch: Phase Two"

jascha_eng — Sun, 21 Jun 2026 01:28:44 +0000

This mostly reads as a comparison between Opus 4.7 and 4.1 it would be more interesting if they reran the experiment against a team of humans with 4.7 and see how much the humans still improve the results today.

New comment by jascha_eng in "Dynamic Workflows in Claude Code"

jascha_eng — Thu, 28 May 2026 17:30:25 +0000

yes I agree with this, more granular going back, letting me interrupt where it went off the rails, or even editing file reads myself etc would be lovely. Ingesting parts of other conversations would also be cool!

New comment by jascha_eng in "Claude Opus 4.8"

jascha_eng — Thu, 28 May 2026 17:25:10 +0000

The benchmark improvements actually look pretty damn nice tho!

New comment by jascha_eng in "Project Glasswing: what Mythos showed us"

jascha_eng — Mon, 18 May 2026 19:00:13 +0000

They kind of have a little diagram explaining the steps I imagine every single step in that to basically be it's own Claude code session.

New comment by jascha_eng in "Anonymous request-token comparisons from Opus 4.6 and Opus 4.7"

jascha_eng — Sat, 18 Apr 2026 18:17:20 +0000

yup its all vibes. And anthropic is winning on those in my book still