News for hackers. Stuff that matters.

mrdw

Re:Apple capped engineers at a $200 daily AI budget on Monday November 09, 2026 @01:26AM (#50421807)
If $200 in API price equivalent it's too low. That's less than Claude Max x5 plan gives. So with $100 plan you'd have more tokens than apple engineer
Re:No Signs of AI Payoff Outside the Tech Sector on Monday October 26, 2026 @04:38AM (#50240855)
I currently work in E-commerce, local online marketplace in Kazakhstan. I scraped all millions of goods in the marketplace, then analyzed top 30k goods with AI. Goal here to find it's supplier in China accurately and estimate weight, similarity, etc. That index gave me huge competitive advantage compared to other sellers and extremely boosting my income. But that cost a lot, because, it's literally like running new Claude Code session for each good of tenth thousands, with custom tools like search in alibaba, viewing reviews, etc.

In marketplaces margin/ROI of goods are distributed like(just for example):

Top 1% goods (P01) - ROI 1000%+

Top 5% goods - ROI 300%+

Top 10% goods - ROI 100%+

Top 20% goods - ROI 50%+

And so on. So such automation helps me to find that Top 1% first

Re:Let's Ditch Google (Verb) on Wednesday September 30, 2026 @01:27PM (#49911852)
source: trust me bro
tests generated by ai, especially which generates each time codex are completely useless and needed only for beautiful "44/44 tests passed"
I disagree. Personally, I would work in company which respects more code/infra quality than coding speed. When you vibe-deliver a lot of features per sprint after some time codebase becomes an unpredictable AI Slop.
What's your grade?
Re:Show HN: I made an emergency page for my family on Wednesday May 27, 2026 @12:07PM (#48296341)
"LLM-summarized" lmao
Re:UK sovereign LLM inference on Friday May 15, 2026 @05:07PM (#48153859)
Yeah, I noticed this library few years ago when checking pypi.org for supply chain attack vulnerability and scanned all libraries. There are a lot of such libraries which you can take over.
Re:UK sovereign LLM inference on Friday May 15, 2026 @07:15AM (#48147184)
btw, you can claim "relax" name instead of "relaxai" on pypi

pypi.org/project/relax is abandoned library, which owner registered via email with expired custom domain, so you can claim this domain and reset owner's account by email.

If everybody presses red button - everybody survives. Ez solution
I noticed another limitation: "An image in the conversation exceeds the dimension limit for many-image requests (2000px). Start a new session with fewer images."

So I can't continue my claude code session I started yesterday.

just use free antigravity subscription with Opus 4.5 and reverse engineered API, and bunch of cheap google accounts
Re:Thank you Google for breaking my YouTube addiction on Sunday May 18, 2025 @08:32AM (#44020908)
Thank you Google for breaking my YT addiction by introducing YT shorts, now its deleted
Re:Are LLMs Random? on Wednesday April 30, 2025 @08:37AM (#43844330)
They should measure for different temperatures, where at 0 it will be the same output every time, but it's interesting to see how results will change for different temperatures from 0.01 to 2. But, I'm not sure if temperature is implemented the same way in all llms
Is only frontend open-source or backend sources are also open? For example Telegram is fully open-source at frontend, but not backend
Cool game. It would be more interesting if the AI-generated picture was a high-quality yet fictional insect species, so you'd have to identify it not by image quality, but by your knowledge of insect species.
Re:A web scraping CLI made for AI that is idempotent on Saturday August 17, 2024 @01:11PM (#41276020)
why so dependent on azure?

"Decoupled architecture with Azure Queue Storage"

"Scraped content is stored in Azure Blob Storage"

"Indexed content is semantically searchable with Azure AI Search"