Max Chernyak

5.7K posts

Max Chernyak banner
Max Chernyak

Max Chernyak

@hakunin

Specialize in software design and DX. VPE at Helios (FinTech). bsky: https://t.co/E07HCRXXb6 | mastodon: @[email protected]

Upstate NY Katılım Aralık 2007
136 Takip Edilen714 Takipçiler
Sabitlenmiş Tweet
Max Chernyak
Max Chernyak@hakunin·
The best kind of programming is not when you get a lot done, but when you painstakingly spend extra time to arrive at something so simple and elegant that it ironically looks effortless and not worthy of any special regard.
English
1
22
77
0
Max Chernyak
Max Chernyak@hakunin·
I scan and shred lots of docs (kids' schoolwork, mail, etc), so wanted a simple way to let AI at them. Vibed this up and been using it happily for a couple months now. Maybe useful to someone else. Contrib welcome! github.com/maxim/ringbind…
English
0
0
0
27
Max Chernyak
Max Chernyak@hakunin·
Tried to have AI read my posts and codify the style, then made several different LLMs try to write as me. The result is unfortunately beyond cringe. Weirdly, simpler models (e.g. Qwen 3.6-plus) came the closest, but it's still painful to read. Hard nope.
English
0
0
0
31
Max Chernyak
Max Chernyak@hakunin·
A new way to recognize AI-built web frontends: they're broken in Safari. As a Safari user, this is getting old fast. I thought AI should help cover more ground?
English
0
0
0
34
plannotator
plannotator@plannotator·
Who is using Codex as an advisor agent to catch bugs??
English
4
1
10
1.5K
Max Chernyak
Max Chernyak@hakunin·
Using plannotator with pi lately (liking it so far). Wrote a seeds file for my Rails app. Code review from Claude: "here're a couple of nits". Code review from Codex: "you're about to break production, because you're doing db:prepare which also runs your seeds".
English
0
1
5
378
Max Chernyak
Max Chernyak@hakunin·
Many think that code is not the goal, working software is the goal, so you don’t need to read code, but I hard disagree. Imo code is the only solid ground we have in all of this non-determinism. AI should produce great code first and foremost, and useful software second.
antirez@antirez

Also note that if you are doing good work with automatic programming, you likely can find a proxy on the number of lines of code of the project: the more lines of code compared to the actual output, the worse. 6k lines of code can be reviewed, 100k? I don't think so.

English
0
0
2
81
Max Chernyak
Max Chernyak@hakunin·
Something's missing in these gstack-style approaches to AI. Why reupload millions of tokens of its own slop back into it? It's in there already. All it needs is what it doesn't have: your data, and some steering to apply its knowledge according to your preferences. Use pi too.
English
0
0
0
28
Max Chernyak
Max Chernyak@hakunin·
@cooperx86 One thing about LG is that sometimes their firmware is shit, but gets fixed later. I got the LG ultragear 5k2k OLED purely for gaming, and it had this awful habit of failing to turn on randomly, but after updates, been solid. At least there’s hope.
English
0
0
0
73
Peter Cooper
Peter Cooper@cooperx86·
Wanted a 32" 6K monitor. Couldn't justify the XDR. Got the LG Ultrafine 6K. Excellent overall *except* color consistency is AWFUL. Blue fringing, lower quarter runs cool. I knew this going in from reviews. It's tolerable for me, but avoid if this would annoy you.
Peter Cooper tweet media
English
3
0
4
1.3K
Max Chernyak retweetledi
Computer ♥ Records
Computer ♥ Records@ComputerLove_·
Cruise the net faster with 2 wheels!
Computer ♥ Records tweet media
English
66
372
5.9K
177.7K
Max Chernyak
Max Chernyak@hakunin·
@jnunemaker Controversial thought. Optimize all workflows for one screen. 14” on the go, plugged in to 1 screen at home (laptop open, tucked behind ext display, brightness off). But I’ve been optimizing for this for many years.
English
0
0
0
232
John Nunemaker
John Nunemaker@jnunemaker·
I can't decide if I want a new m5 maxed out or an air + light second screen for on the go + a beefy server sitting somewhere on the internet.
English
6
0
9
4.5K
Max Chernyak
Max Chernyak@hakunin·
@cooperx86 That’s a superpower. I’ve always been ok with sleeping itself, but could never bring myself to bed on time, so end up chronically with 5-6 hour nights, because kids have school.
English
0
0
0
15
Peter Cooper
Peter Cooper@cooperx86·
@hakunin I’m a very deep sleeper. The time is reducing as I get older though which is actually quite welcome!
English
1
0
1
182
Peter Cooper
Peter Cooper@cooperx86·
iOS 26 has added a notification that looks just like the high heart rate notification except for something good 😂 Red, heart, “high”.. panic! Oh no, it’s just saying I had a good sleep. Turning it off.
Peter Cooper tweet media
English
2
0
1
518
matt swanson 😈
matt swanson 😈@_swanson·
@zeeg I also don't understand how multi tenant web app A is supposed to allow agentic tool calls to web apps B...Z without it...I'm just supposed to have thousands of CLIs on my compute?
English
1
0
2
1.1K
David Cramer
David Cramer@zeeg·
I really wish folks would step back and consider the anti-value of broadcasting uninformed opinions. "mcp is bad" usually is generally 1) doesnt work with large oragnizations, 2) doesnt understand context steering, 3) doesn't understand the value of auth cra.mr/mcp-skills-and…
nisten🇨🇦e/acc@nisten

MCP servers are just bad skills files. On that note, wondering why not just make a standard openAPI endpoint with regular discovery ability (which MCP still doesn't have lol) and feed the bot the skill tree AND backend via that.....

English
16
4
98
55.1K
Peter Cooper
Peter Cooper@cooperx86·
@petergyang On 14 now and while it’s a fantastic form factor the screen is severely restricted for modern agentic use cases for me. I use my iPad Pro as a second screen. Considering 16 next time as I rarely travel.
English
2
0
1
936
Peter Yang
Peter Yang@petergyang·
I want to buy a new personal MacBook Pro - I’m still on M1 😅 Apparently Apple has both M5 and M6 coming out? Do you all prefer the 14 or 16 inch? Use case is vibe coding and video editing
English
338
8
627
181.4K
DHH
DHH@dhh·
Kevin: "I just raced Claude and Kimi K2.5 against that bug that Ryan was talking about. K2.5 fixed it in 21s. Claude took just over a minute to make the plan, then about 2 minutes to execute on it. Both had the same fix, though." (K2.5 is now my main driver. Opus just backup.)
English
169
106
3.2K
274.6K
DHH
DHH@dhh·
@kmckelvin I run Kimi off @opencode Zen, which uses @FireworksAI_HQ for inferrence. So all of the computing is happening in the West. Like running any form of open source code.
English
9
5
219
17.8K
Max Chernyak retweetledi
Pedram Amini
Pedram Amini@pedramamini·
Day two of my entire Github being wiped off the face of the planet with zero explanation. @github why can't I get any response from your support team! I'm left with no recourse here.
English
5
8
46
17.1K

Keşfet

@plannotator @cooperx86 @jnunemaker @_swanson @zeeg @elonmusk @BarackObama @taylorswift13