HackerTrans
TopNewTrendsCommentsPastAskShowJobs

mihau

no profile record

Submissions

Korpo Clicker

app.korpoclicker.pl
1 points·by mihau·2 mesi fa·0 comments

Simple Self-Distillation

github.com
3 points·by mihau·3 mesi fa·0 comments

State of the Art Speech Recognition with MAI-Transcribe-1

microsoft.ai
2 points·by mihau·3 mesi fa·0 comments

Voxtral TTS

mistral.ai
3 points·by mihau·3 mesi fa·0 comments

LLM SQL Benchmark

sql-benchmark.nicklothian.com
2 points·by mihau·3 mesi fa·0 comments

Microsoft.ai

microsoft.ai
4 points·by mihau·3 mesi fa·1 comments

Cherri – programming language that compiles to an Apple Shortuct

github.com
355 points·by mihau·4 mesi fa·74 comments

Show HN: Record manual QA flows, get E2E test code that fits your repo

quretests.com
19 points·by mihau·4 mesi fa·2 comments

The Wrong Abstraction

sandimetz.com
2 points·by mihau·4 mesi fa·0 comments

Can Good Writing Be Generative?

arxiv.org
3 points·by mihau·4 mesi fa·0 comments

Markdown UI – rendering interactive widgets in plain Markdown

github.com
3 points·by mihau·4 mesi fa·0 comments

Sandcat – Docker and dev container setup for securely running AI agents

github.com
8 points·by mihau·4 mesi fa·0 comments

Stacksort

gkoberger.github.io
1 points·by mihau·4 mesi fa·0 comments

Mq – a command-line tool that processes Markdown using a syntax similar to jq

github.com
4 points·by mihau·5 mesi fa·0 comments

Peon-ping – Your Peon pings you the instant Claude Code finishes

peon-ping.vercel.app
2 points·by mihau·5 mesi fa·0 comments

GLM-5 by Z.ai was released

twitter.com
4 points·by mihau·5 mesi fa·1 comments

CLI – hooks into your Git workflow to capture AI agent sessions

github.com
1 points·by mihau·5 mesi fa·0 comments

[untitled]

1 points·by mihau·6 mesi fa·0 comments

Weather Spark – The Weather Year Round Anywhere on Earth

weatherspark.com
2 points·by mihau·6 mesi fa·0 comments

Everything you might have missed in Java in 2025

jvm-weekly.com
11 points·by mihau·6 mesi fa·1 comments

comments

mihau
·5 mesi fa·discuss
I agree that Claude Code is buggy as hell, but:

> Let's not talk about the feature parity with other agents.

What do you mean feature parity with other agents? It seems to me that other CLI agents are quite far from Claude Code in this regard.
mihau
·5 mesi fa·discuss
same
mihau
·8 mesi fa·discuss
@simonw wen pelican
mihau
·8 mesi fa·discuss
FYI it's behind a feature flag (aka experiment). Just in case you rarely use experiments in DevTools:

DevTools -> Settings (cog, top right) -> Experiments -> Search for "Protocol Monitor" -> Check the checkbox
mihau
·9 mesi fa·discuss
why? M3 Ultra already had 800 GB/s (6400 gbps) memory bandwidth
mihau
·9 mesi fa·discuss
It feels to me that the OP on the forum expects this to work: "read this existing function, then read my mind and do stuff" (probably followed by "do better").

It still takes a lot of practice to get good at prompting, though.
mihau
·9 mesi fa·discuss
> you can overengineer your prompt to try get them to ask more questions

why overengineer? it's super simple

I just do this for 60% of my prompts: "{long description of the feature}, please ask 10 questions before writing any code"
mihau
·9 mesi fa·discuss
Kinda off-topic but I love the quality of images used in this article. Even memes are HD. Quite rare as people tend to use whatever meme creator provides - super-pixelated, low-quality source image with text slapped on top.
mihau
·9 mesi fa·discuss
Yep, it's just bad. Physics feels totally off.
mihau
·10 mesi fa·discuss
https://archive.is/lJuT0
mihau
·10 anni fa·discuss
Great article :). Cal Newport has some reaaaallly important ideas. I encourage you to read his book Deep Work in which he explains how to create deep, complex stuff.