Hacker Newsnew | past | comments | ask | show | jobs | submit | raffkede's commentslogin

Skills is actually what also Claude code uses internally, it's cool because the llm will load the whole context on how to use it only on demand and keeps the context cleaner.

.99 USD with the Kimi 2k5 deal

Didn't test this but the Design is too close to the original page

I think the slow feeling is a UI thing in codex

I realize my comment was unclear. I use codex the CLI all the time, but generally with this invocation: `codex --full-auto -m gpt-5.2`

However, when I use the 5.2codex model, I've found it to be very slow and worse (hard to quantify, but I preferred straight-up 5.2 output).


I would be surprised if Copilot is even close to that

And if I watch a movie it will pop up after 5 minutes

A Calculator won't increase your creativity directly but it will free resources that you can allocate to creativity!


Securikett | Senior Software Developer | Krems an der Donau (near Vienna), Austria (ONSITE)

Senior Fullstack Developer (1 UI/UX-focused role, 1 general role)

We are Securikett (https://www.securikett.com/), a 100% family-owned company in product protection and anti-counterfeiting. Beyond security labels, we are building a SaaS platform extending physical product protection into the digital world. The new system is already in production and will replace legacy systems over the next year. Software is a top strategic priority.

You’ll work in a small team (5 devs + 2 contractors) and own features end to end.

What you’ll do: build and evolve a SaaS product for international customers; React frontend (S3 + CloudFront); Java “lambdalith” backend on AWS Lambda (Java 25); UI/UX role: push UX quality and frontend maintainability.

What we’re looking for: 5+ years production experience; strong engineering fundamentals; Java + JavaScript (other languages ok); AWS experience a strong plus (everything as CDK IaC); comfortable working autonomously.

Tech / How we work: React, Java on AWS Lambda, serverless, strong focus on automated tests. “No Ops”, highly automated infrastructure, direct deploys to production, close collaboration with company owners.

Location: ONSITE in Krems an der Donau / Wachau region (near Vienna). Hybrid setup: ~40–60% home office after onboarding.

Compensation: €70k–€120k depending on experience. Austria offers strong public benefits (free childcare, kindergarten, universities), increasing effective compensation.

Apply: Email your CV and a short personal intro (no fluff) to raffael.kainersdorfer+hn@[companydomain].com I’m Raffael Kainersdorfer, Senior Developer / Tech Lead, personally involved in hiring.


Totally wild to see Krems an der Donau here haha :)


Everyone shitting on this without looking should look at the creator, and/or try it out. I didn't really dive in but its extremely well integrated with a lot of channels, to big thing is all these onnectors that work out of the box. It's also security aware and warns on the startup what to do to keep it inside a boundary.


The creator is a big part of what concerns me tbh. He puts out blog posts saying he doesn’t read any of the code. For a project where security is so critical, this seems… short sighted.


Seems to be the first model that one-shots my secret benchmark about nested SQLite and it did it in 30s,


Out of interest. Does it one shot it every time?


Will try again just tried once in the phone a few hours ago, other models were able to do quite a lot but usually missing some stuff this time it managed nested navigation quite well, lot of stuff missing for sure I just tested the basics with the play button in AI studio


It seems to be that first impression that makes all the difference. Especially with the randomness that comes with llms in general. which maybe explains the 'wow this is so much better' vs the 'this is no better than xxx' commments littered throughout this whole parent post.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: