Hacker News Reader

loreyon Volkswagen Just Built an EV That Can Go Nearly 900 Miles on a Charge

They did something very similar 20+ years ago called XL1 to achieve 1l/100km, never made it into production. https://de.wikipedia.org/wiki/VW_XL1

loreyon How I advertise malicious software on Google Ads

Had the same issue. In the end, getting pulled into some Google Ads sales funnel and them solving the issue through internal tickets helped. Spent $1000 on useless ads in the process, but have an "unlocked" account now. Really disappointing customer experience.

loreyon Pareto Front

Found this to display the optimal LLM choice while building evalry. It's such a useful tool, not only for thinking about it, but for visualization, too.

Example: Which LLM gives me the best ELI5 explanations for a given price. https://evalry.com/benchmarks/explain-like-i-m-5-321

loreyon Interview with Boris Cherny [video]

What I read from that is that there's a chance claude gets worse using a claude.md and that's a usability issue on their side.

loreyon Hey Nico, you didn't vibe code your data room but stole it from Papermark

Their response:

> The team that made dataroom has stated that they did not use any of papermark’s code and that dataroom was made from scratch with inspiration from existing document sharing softwares, and that this post’s allegations of us stealing code are false. [...]

The screenshots clearly show they copied whole pages verbatim, both design and texts. The founder, Nico Laqua, basically responding with "we didn't copy _code_" and not taking any responsibility says a lot about his and his company's moral code. It might not be enough to get sued. That doesn't make it right.

https://x.com/nico_laqua/status/2070158170937581951

loreyon Without benchmarking LLMs, you're likely overpaying

That is not what the article argues.

loreyon Without benchmarking LLMs, you're likely overpaying

Haha, very true. Exactly as described in the article.

loreyon Without benchmarking LLMs, you're likely overpaying

This is true with one caveat.

In most cases, e.g. with regular ML, evals are easy and not doing them results in inferior performance. With LLMs, especially frontier LLMs, this has flipped. Not doing them will likely give you alight performance and at the same time proper benchmarks are tricky to implement.

loreyon Without benchmarking LLMs, you're likely overpaying

This is a very good point. When I came in, the founder did a lot of evaluation based on a few prompts and with manual evaluation, exactly as described. Showing the results helped me underline the fact that "works for me" (tm) does not match the actual data in many cases.

loreyon Without benchmarking LLMs, you're likely overpaying

Doesn't this depend a lot on private vs company usage? There's no way I could spend more than a few hundreds alone, but when you run prompts on 1M entities in some corporate use case, this will incur costs, no matter how cheap the model usage.

loreyon Without benchmarking LLMs, you're likely overpaying

It's not you, it's the HN hug of death. There's so much load on the server, I'm barely able to download the redis image I need for caching...

loreyon Without benchmarking LLMs, you're likely overpaying

Thanks. Will take a look.

loreyon Without benchmarking LLMs, you're likely overpaying

Depends on your remaining budget ;)

loreyon Without benchmarking LLMs, you're likely overpaying

I've skipped that in the article, but absolutely!

loreyon Without benchmarking LLMs, you're likely overpaying

Fixed, thanks. Not a native speaker.

loreyon Without benchmarking LLMs, you're likely overpaying

That's interesting. Similarly, we found out that for very simple tasks the older Haiku models are interesting as they're cheaper than the latest Haiku models and often perform equally well.

loreyon Without benchmarking LLMs, you're likely overpaying

Pushed a fix. Could you check, please?

Any resources you can recommend to properly tackle this going forward?

loreyon Without benchmarking LLMs, you're likely overpaying

Will fix, thanks :)

loreyon Without benchmarking LLMs, you're likely overpaying

Totally agree with your point. While I can't say specifically, it's a traditional (German) business he's doing vertically integrated with AI. Customer support is really bad in this traditional niche and by leveraging AI on top of doing the support himself 24/7, he was able to make it his competitive edge.

loreyon Without benchmarking LLMs, you're likely overpaying

This went straight to prod, even earlier than I'd opted for. What do you mean?

loreyon Without benchmarking LLMs, you're likely overpaying

Appreciate the feedback, will work on that.

loreyon Without benchmarking LLMs, you're likely overpaying

You're right. We did a few use cases and I have to admit that while customer service is easiest to explain, its where I'd also not choose the cheapest model for said reasons.

loreyon Without benchmarking LLMs, you're likely overpaying

Yes, absolutely. This aligns with what we found. It seems to be necessary to be very clear on scoring (at least for Opus 4.5).

loreyon Everything as code: How we manage our company in one monorepo

Very interesting points. Would you mind sharing a few examples of when cherry-picking is necessary and why atomic changes are a lie?

I'm using a monorepo for my company across 3+ products and so far we're deploying from stable release to stable release without any issues.

loreyon Show HN: Free Logo API – logos for any company or domain

Yes, there's the <link rel="icon"...> an other specific tags providing far more options than the old favicons.

For anyone interested: https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/...

loreyon Show HN: Free Logo API – logos for any company or domain

Thanks, this is how I feel about this, too.

loreyon Ask HN: What Are You Working On? (Nov 2025)

Spent the last three months building a competitor/lookalike ML model + API. Started using plain embedding similarity and quickly realized you end up with similar noisy results as ocean.io. Ended up using similarity learning which works quite well with little data. Launched this as an API and small web app. Hardest part right now is to fend off scrapers honestly.

Examples:

- YC: https://markets.apistemic.com/companies/y-combinator-goaq9

- uber: https://markets.apistemic.com/companies/uber-com-ojj2j

- Anthropic: https://markets.apistemic.com/companies/anthropicresearch-yx...

Try it for any company here: https://markets.apistemic.com

loreyon Vibe coding as a VC

I don't understand why so many VCs fall for "not invented here". Vibe-coded or not, this is just another in-house solution, inferior and more expensive than most out-of-the-box products already out there.

loreyon I algorithmically donated $5000 to Open Source

Had to modify the title slightly to pass validation. Also "How [I algo...]" got removed, did not know that happens automatically.

loreyon Experimental library for scraping websites using OpenAI's GPT API

Why GPT-based then? There are libraries that do this: You give examples, they generate the rules for you and give you a scraper object that takes any html and returns the scraped data.

Mine: https://github.com/lorey/mlscraper Another: https://github.com/alirezamika/autoscraper