HackerTrans
TopNewTrendsCommentsPastAskShowJobs

lukax

889 karmajoined 15 years ago

Submissions

Vite Vulnerable to Arbitrary File Read via Vite Dev Server WebSocket

github.com
2 points·by lukax·3 months ago·1 comments

Soniox: Real-time transcription in 60 languages

soniox.com
2 points·by lukax·5 months ago·1 comments

Another user's pCloud setup is visible in my pcloud drive

old.reddit.com
12 points·by lukax·5 months ago·3 comments

Gemini CLI bot infinite loop

github.com
3 points·by lukax·6 months ago·0 comments

Gemini Live Speech Translation

blog.google
2 points·by lukax·7 months ago·0 comments

Did Nvidia Just Prove There Is No AI Bubble

planetearthandbeyond.co
21 points·by lukax·7 months ago·13 comments

Tile Language: DSL for High-Performance GPU/CPU/Accelerators Kernels

github.com
6 points·by lukax·9 months ago·1 comments

Tell HN: Opt-out of LinkedIn training content creation AI models

13 points·by lukax·10 months ago·7 comments

comments

lukax
·2 days ago·discuss
Could it be that users of Pi are more senior and know better how to prompt and that's why the pass rate is higher?
lukax
·2 days ago·discuss
Yes, this is a known issue. A significant amount of Edit tool calls fails in Pi witg newer models.

https://lucumr.pocoo.org/2026/7/4/better-models-worse-tools/
lukax
·4 days ago·discuss
Well, Microsoft just started offering Kimi K2.7 through Copilot hosted on Azure.

https://github.blog/changelog/2026-07-01-kimi-k2-7-is-now-av...

Cursor Composer 2 and 2.5 are also fine tunes of Kimi K2.5

It looks like politics don't matter when it comes to economics.
lukax
·11 days ago·discuss
It's refreshing to see a tech article that isn't about AI. It feels like 5 years ago.
lukax
·12 days ago·discuss
NUMA can cause really crappy performance. We deployed a Go based LLM gateway in Kubernetes deployed on a server with hundreds of CPU cores. We didn't explicitly set GOMAXPROCS so Go runtime scheduled goroutines over different CPUs and it constantly used 200% CPU and GC was causing latency spikes. Then we set GOMAXPROCS 8 and all performance issues went away. Until recently Kubernetes didn't work well with NUMA.
lukax
·19 days ago·discuss
Well, that's how data for training LLMs is scraped.
lukax
·last month·discuss
Zero data retention policies.
lukax
·last month·discuss
There is also Retrotick.

https://retrotick.com/

It simulates x86 (win32 and win16) and implements Windows APIs in javascript and renders window frames with DOM and contents with canvas (e.g. GDI translates to browser canvas operations). A lot of programs run already but a lot of APIs are not yet implemented.

I successfullt spent a few days extending it to run a Click & Create based game from my childhood.
lukax
·2 months ago·discuss
Huawei Ascend AI Accellerators. DeepSeek V4 model architecture was optimized for Chinese hardware.
lukax
·2 months ago·discuss
That's just not true. Let's say you have a form validation library with a public api that supports custom validators Validate(name string, value string) bool. Then you decide that your validator now needs to make an HTTP request. This request needs context so that tracing is propagated and needs to return (bool, error) so that error is propagated up instead of silently ignoring it or logging it and returning false. This is coloring. You can use context.Background the same way you can use blocking in other languages. It just doesn't feel right and it breaks things.
lukax
·2 months ago·discuss
I wonder what would happen if they used Kimi 2.5 directly instead of Cursor Composer 2.5. Composer is a fine tune of Kimi. Probably they didn't want to test "Chinese" models.
lukax
·2 months ago·discuss
It looks like migration to Azure is not going very well

https://news.ycombinator.com/item?id=45517173
lukax
·2 months ago·discuss
They are migrating from their own datacenters to Azure
lukax
·3 months ago·discuss
See the implementation of Python's math.isclose

https://github.com/python/cpython/blob/d61fcf834d197f0113a6a...
lukax
·3 months ago·discuss
You generally want both relative and absolute tolerances. Relative handles scale, absolute handles values near zero (raw EPSILON isn’t a universal threshold per IEEE 754).

The usual pattern is abs(a - b) <= max(rel_tol * max(abs(a), abs(b)), abs_tol) to avoid both large-value and near-zero pitfalls.
lukax
·3 months ago·discuss
Do you really need to roll your own NIO HTTP server? You could just use Jetty with virtual threads (still uses NIO under the hood though) and enjoy the synchronous code style (same as Go)
lukax
·3 months ago·discuss
Rust wasm ecosystem also needs a lot of crates to do anything useful, a lot of them unmaintained.
lukax
·3 months ago·discuss
Combine that with CVE-2025-24010 and any website was able to read any file on developers' computers.

https://github.com/advisories/GHSA-vg6x-rcgg-rjx6
lukax
·4 months ago·discuss
Looks nice but still a bit sad that Flutter is used instead of something native given that they don't need the app to be cross-platform.

Well, even Microsoft uses React Native for a lot of Windows-only apps.
lukax
·4 months ago·discuss
Sorry, I commented too soon. Did you also try Soniox? Why did you decide to use Deepgram's Flux (English only)?