@codeshaunted

compilers by day @boundaryml reverse engineering by night @ https://xitter.cf/t.co/STAHbkiMjR helped build @darkflameuniv

 ~/
Joined April 2017
  • Tweets1.5K
  • Following199
  • Followers1.4K
166Media
in the next version of Porffor, Test262 compliance leaps to 96%! this marks a milestone where Node compat and performance finally become major blockers
"porffor passes 96.05% of test262" with a graph showing progress over time

ALT "porffor passes 96.05% of test262" with a graph showing progress over time

14
16
3
260
10,085
World of Goo 2 is Steam’s Deal of the Day, and is 50% off for the next week! We’re also putting our other games on sale, in case you know someone who could use a nice cozy fireplace simulator, or management-themed programming puzzles! store.steampowered.com/sale/…
2
41
2
266
4,172
37
1,357
49
28,653
331,052
thx jev
Diogo Almeida
@CompleteSkepticOct 14
Replying to @CompleteSkeptic
Types Beyond Decisions: box We’re also releasing a new type that is somewhere in the gray area of decisions: a single bounding box. We warned you there’s a reason we call them System One Models! (only on jev-preview)
1
2
4
523
Purchased.
217
7,321
260
95,776
1,067,421
avery retweeted
arn@anemoiac_Oct 13
reading item descriptions as a new player and I see "active"
230
2,716
166
57,931
1,036,590
We realize it’s been a while since we’ve posted anything, but we’re still here, still plugging away. We hope we have something to share soon… Stay Warm In There!
25
48
8
653
6,652
avery retweeted
Snoz@snozzzzzzOct 12
BREAKING: ASSHOLE roommate steals ketchup from fridge and keeps it in his room
1
1
1
77
avery retweeted
Vaibhav Gupta
@vaibcodeOct 7
Replying to @codeshaunted
@codeshaunted just converted me to: terse good bye to @dillon_mulroy's /bro skill. we had a good run 🥲
2
2
15
1,000
20
132
14
3,925
109,091
Gemini in the Waymo wasn't willing to reveal its tool list, but we convinced it to tell us .. in the form of a story!
20
16
6
431
101,685
avery retweeted
anna
@meowkoteeqOct 4
[IntelliSense Disclosure] I have to admit, I used IntelliSense and syntax highlighting to assist me in with this project. I struggle to remember all the symbols of the standard library, so these tools come in handy and speed up my development flow. Hope for your understanding.
24
124
12
2,754
137,831
computah, uhh put... cuphead into... uhh... fallout. computah... do
118
1,832
51
40,428
446,340
avery retweeted
sam@sxlijinSep 30
1
1
38
this is one of the most dogshit websites i've seen, and they're worth $98 BILLION. why are there like 15 diff hues of purple, blue and... green? why does the video look like it was made in Microsoft Powerpoint?
2
1
10
1,312
A bad specific example I won't name. This benchmark is on their website as "factorial".
// Benchmark: Large number computation // Measures numeric computation with overflow handling const ITERATIONS = 100000000; let sum = 0; const start = Date.now(); for (let i = 0; i < ITERATIONS; i++) { // Simulate factorial-like accumulation pattern sum = sum + (i % 1000); } const elapsed = Date.now() - start; console.log("accumulate:" + elapsed); console.log("sum:" + sum);

ALT // Benchmark: Large number computation // Measures numeric computation with overflow handling const ITERATIONS = 100000000; let sum = 0; const start = Date.now(); for (let i = 0; i < ITERATIONS; i++) { // Simulate factorial-like accumulation pattern sum = sum + (i % 1000); } const elapsed = Date.now() - start; console.log("accumulate:" + elapsed); console.log("sum:" + sum);

Oliver Medhurst
@CanadaHonkSep 24
I avoid dumping on other projects as I try to be positive but seeing so many vibe coded TS compilers flex microbenchmarks saying they are Nx faster than Node as headline figures on their website is incredibly misleading and makes me want to make a serious comparison instead
20
4
3
252
34,626
I avoid dumping on other projects as I try to be positive but seeing so many vibe coded TS compilers flex microbenchmarks saying they are Nx faster than Node as headline figures on their website is incredibly misleading and makes me want to make a serious comparison instead
11
3
1
145
36,388
avery retweeted
Ray
@raysan5Sep 22
How is it possible to get that level of open source support while many other projects like raylib struggle to get enough finance to even pay a developer minimal salary? I don't know... 😞
DHH
@dhhSep 22
Thrilled to announce @alibaba_cloud as a Founding Corporate Patron for the Omacom Foundation! $3 million in funding, collaboration on Omarchy China, and bringing Omarchy to the newly announced Qwen Book. Agentic computers need a native agentic OS! omarchy.org/news/2026/09/ali…
299
77
18
1,920
107,659
In the last few years I've traveled to many places. I've been to Seoul, Tokyo, Osaka, Dublin, Paris, Seattle, SF, LA, San Diego. But today I find myself in a small village in the middle of nowhere, hidden deep in Romanian plains. This is where I come from. Raised in a house with no running water, sinks, or indoors toilets, by people working these lands. When I was still a kid my dad moved to Spain to work in construction and brought us to settle there. But things didn't go particularly well in Spain either after the 2007 housing crash. And yet here I am now, writing fucking compilers and VMs for agents in Rust. I overcame poverty and helped my family purely by writing insane amounts of code. "You can just do things" has never been truer than in tech. – at Tritenii de Jos, România
12
8
2
191
5,248
a lot of the slowness in modern JS is having to leave the JIT. to be fast, built-in functions have to be implemented multiple times because moving from JIT execution to engine world is opaque. (this is also a part of why modern JS engines are so huge/complex)
Dmitriy Kovalenko
@neogoose_btwSep 15
What I realized that even experienced JS devs don’t fully understand how JS is executing. We all hear word JIT, byte code optimizations, but it is very EXPENSIVE procedure to compile javascript code so absolute majority of the code outside of hot loop stays interpreted. JavaScript engines byte code is also not the real native CPU code that they can freely execute. This is another VM inside your runtime that executes optimized either stack/register virtual code that also needs an additional virtualization mechanism to run on the CPU. 1) Not all the code is possible to JIT compile 2) When you compile JIT code you have to store in runtime memory which leads to increased RSS
5
7
101
8,697
Load more