HNHacker News
TopNewBestAskShowJobs

alexeiz

966 karma · joined February 20, 2011

submissionscomments
alexeiz··on Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows
I got Muse Glimmer to say "Drive the car, walk yourself." The logic is unbeatable.
alexeiz··on Muse Code and Muse Spark 1.2
Muse code is rough around the edges. But combined with almost free model (muse spark contributor) it's actually pretty good. I think it's on the same level as grok build.
alexeiz··on Google releases Gemma 4 open models
It's time to do "frog on a skateboard" now.
alexeiz··on GitHub is down again
There are new slides? Here goes the rest of my work day.
alexeiz··on GitHub is down again
You're absolutely right! Sorry I deleted your database.
alexeiz··on 175K+ publicly-exposed Ollama AI instances discovered
Fear not, we have Clawdbot! (openclaw whatever) You can put your naked ass on the internet in seconds now.
alexeiz··on Anthropic blocks third-party use of Claude Code subscriptions
> The unfortunate truth is that the engineering in opencode is so far ahead of Claude Code.

If only Claude Code developers had access to a powerful LLM that would allow them to close the engineering gap. Oh, wait...

alexeiz··on Show HN: Mgrep – A Semantic, Multimodal Grep
Log in and send your data to some unknown company just to do grep. Thanks but no thanks.
alexeiz··on DeepSeek-v3.1-Terminus
It says the best "coding index" is held by Grok 4 and Gemini 2.5 Pro. Give me a break. Nobody uses those models for serious coding. It's dominated by Sonnet 4/Opus 4.1 and GPT-5.
alexeiz··on The Open-Office Trap (2014)
This reminds me something. Ah yes, Congress. They'd rather be in power in a failed country.
alexeiz··on I forced every engineer to take sales calls and they rewrote our platform
Here's what I see:

* PM asking a client: "Do you want beautiful logging, metrics and shit?" - "Hell yeah we want logging, metrics and shit!"

* PM telling devs: implement beautiful logging, metrics and shit

* Devs spend half a year implementing all that shit

* The client looks at the final product: "What is this shit? We need a freaking green checkmark!"

alexeiz··on Font Comparison: Atkinson Hyperlegible Mono vs. JetBrains Mono and Fira Code
Coming from Commit Mono, Atkinson looks a bit unusual. But I think I can get used to it. I think the comparison to Fira Code is valid, because in the terminal Atkinson looks almost like Fira Mono, but better. Since I usually sit a meter away from the screen, I can appreciate the extra legibility of this font.

Also, it's great that it's available as a Nerd variant. It makes it super easy to install on Linux with Embellish.

alexeiz··on I Used Arch, BTW: macOS, Day 1
For some time after the introduction of the M1 CPU Macbooks provided the best battery life experience. So it would make sense to use a Macbook if you wanted your laptop to work more than 4-5 hours unplugged. But now Intel Core Ultra series 2 and AMD Strix Point laptops are almost as good. Sure you won't get 20 hours of battery life, but you get about 10-15 hours and it's already good enough. Performance of Intel/AMD CPUs are also good enough: about M2 performance-wise. Sure they are a couple of generations behind Apple, but for most tasks it doesn't matter. Linux compatibility is also getting there. You can choose a Thinkpad if you want to be sure that Linux runs on it. But even many consumer-grade laptops are supported but Linux thanks to recent kernel updates. Essentially what I'm saying is that if you want a good laptop experience and you like Linux, you don't have to settle for Macbook anymore.
alexeiz··on Writing Toy Software Is a Joy
Most difficulty and time estimates are random. Chess engine 2 days? Like a really shitty chess engine maybe. Regex engine 4/10, 3 days. Again a really shitty regex engine. What's the point of doing a bunch of projects in a really shitty way?
alexeiz··on Ask HN: I'm an MIT senior and still unemployed – and so are most of my friends
Don't just complain. Put your side projects on github and show them off.
alexeiz··on Show HN: I built a tool to find devs based on code, not LinkedIn titles
Found whopping 10 people in NYC with Python and JavaScript experience.
alexeiz··on 21st Century C++
He's looking to sell Circle. That must be the reason he's not open sourcing it.
alexeiz··on macOS 15.2 breaks the ability to copy the OS to another drive
> Second person told me to buy a new Mac with 2 tb ssd and restore to that.

Or you can just upgrade the SSD as you would do with any normal laptop. Oh, wait...

alexeiz··on Vanilla OS 2: an immutable distribution to run all software
Vanilla OS is crap. I wanted to try in in a VM and it wouldn't even install. First, it complained that the disk size wasn't enough. Then it complained that it absolutely required UEFI. Then the installer just failed with a cryptic error showing a line of Golang code.
alexeiz··on I'm tired of fixing customers' AI generated code
I had a related episode at work when my coworker asked me why his seemingly trivial 10 line piece of code was misbehaving inexplicably. It turned out he had two variables `file_name` and `filename` and used one in place of another. I asked him how he ended up with such code, he said he used copilot to create it. Using code from a generative AI without understanding what it does is never a good idea.
alexeiz··on The workers have spoken: They're staying home
> boss that saw my desk and started complaining to HR that it was too big

Wow, your boss is a real dickhead.

alexeiz··on Senior employees, ordered back to the office, are jumping ship
> spun out of a much larger corporation

> ZBB

Microsoft?

alexeiz··on Stupid Slow: The Perceived Speed of Computers
Software products these days are being written quick and dirty to be released as fast as possible to capture the market. Nobody has time to create quality software because by the time you finish your product, your users will already be using your competitor's product which may be horribly slow and buggy and a memory hog, but it was released a year earlier than your product. So there you go. "Release fast, improve later" is the motto of the software companies today.
alexeiz··on What is the "God Mode" folder in Windows 10, and how do I enable it? (2022)
Unfortunately "pin to quick access" doesn't do anything for this "all tasks" folder.
alexeiz··on MTA banned from using facial recognition to enforce fare evasion
What facial recognition are you talking about when people jump over the turnstiles and walk into the open gates without any regard for police officers that stand nearby and just watch this happen? MTA is a joke. I keep wondering why am I even paying for the subway?
alexeiz··on Overview: What are Cpp2 and cppfront? How do I get and build cppfront?
That's the problem, it's a single cpp file because it's built manually. The source code is spread among many hpp files for no reason other than being able to include everything in that single cpp file. The project organization of cpp2 is abysmal.
alexeiz··on Best engineering interview question I've gotten
That's why employers need to spend more time convincing potential candidates that the role they hire for is actually worth the trouble of going through this kind of coding interview test.

In the past I've encountered companies that would just throw a multi-hour coding test at you without even giving you any idea of the role they are hiring for. If it's not worth employer's time to get the candidate interested in a position, the position is not worth applying for.

alexeiz··on Terminal Latency
If I use a terminal with 30msec latency to connect to a host with another 30msec network latency, it's now 60msec - pretty noticible by an average human being. That's why reducing latency where you can to the minimal possible value is important.
alexeiz··on Terminal Latency
Zutty is useful for only one thing: passing VTTEST. I've tried it and it's unusable as the terminal you want to rely on to do everyday work.
alexeiz··on Do we need swap on modern systems? (2017)
A lot of swap on servers is bad. A little bit is fine though and can actually be helpful in certain scenarios. I've seen servers with 256GB of RAM configured to have <1GB of swap, so in case of a runaway process the swap fills up very quickly and doesn't delay the OOM killer, but still it helps the Linux memory manager to run more optimally.
Page 1 of 14Next →