Harshil

3.8K posts

Harshil banner
Harshil

Harshil

@hoppingturtles

nerd, gamer, programmer, bitcoiner - all in one. // Studying aerospace @NCState

United States Katılım Ekim 2015
363 Takip Edilen193 Takipçiler
Dan
Dan@KettlebellDan·
Looking to get back into drinking. Any recommendations?
English
337
6
268
26K
Harshil
Harshil@hoppingturtles·
@SpaceX 34 landings is absolutely insane
English
0
1
6
180
SpaceX
SpaceX@SpaceX·
Falcon 9’s fleet leader booster completes its 34th launch and landing
English
527
1.4K
9K
35.1M
Harshil
Harshil@hoppingturtles·
@karpathy that's why I can't take LLMs seriously, they need to be opinionated and stand their ground unless the truth says otherwise
English
0
0
1
15
Andrej Karpathy
Andrej Karpathy@karpathy·
- Drafted a blog post - Used an LLM to meticulously improve the argument over 4 hours. - Wow, feeling great, it’s so convincing! - Fun idea let’s ask it to argue the opposite. - LLM demolishes the entire argument and convinces me that the opposite is in fact true. - lol The LLMs may elicit an opinion when asked but are extremely competent in arguing almost any direction. This is actually super useful as a tool for forming your own opinions, just make sure to ask different directions and be careful with the sycophancy.
English
1.7K
2.4K
30.9K
3.2M
Will McGugan
Will McGugan@willmcgugan·
How can Thai food be so consistently good? Fancy restaurant or food court, it’s always fabulous. Good job I have a decent spice tolerance, because they love a bit of 🌶️
Will McGugan tweet mediaWill McGugan tweet media
English
3
0
21
1.5K
Deepak
Deepak@dipakkr_·
Just got accepted into @AnthropicAI’s Claude for Open Source Program 🎉 Claude Max 20x free for the next 6 months 🚀 Can’t wait to build with this. Thanks team @AnthropicAI @claudeai 🙏
Deepak tweet media
English
3
0
7
1.2K
Saurabh Arora
Saurabh Arora@iSaurabhArora·
@MishaalRahman If this is the path forward, consider disabling programmatic checks for developer options. I have them enabled & have to turn it off whenever I want to use a banking app. Now with a 24 hour wait period to turn it back on, it's going to be annoying AF. It's just punishing devs
English
2
0
8
333
Mishaal Rahman
Mishaal Rahman@MishaalRahman·
📢 Important update on sideloading on Android It’s of course here to stay. The team’s been listening carefully to feedback from power users who want a way (apart from ADB) to install apps that don’t go through developer verification. Today, we have more details on the advanced flow that gives you this option. Read the post👇 for more info goo.gle/advance-flow
Mishaal Rahman tweet media
English
200
112
1.1K
156.3K
Harshil
Harshil@hoppingturtles·
@charliermarsh this can either be a really good thing or really bad. There's no in between. I hope it's the former.
English
0
0
0
44
Charlie Marsh
Charlie Marsh@charliermarsh·
We've entered into an agreement to join OpenAI as part of the Codex team. I'm incredibly proud of the work we've done so far, incredibly grateful to everyone that's supported us, and incredibly excited to keep building tools that make programming feel different.
English
286
145
3.2K
477.5K
Harshil
Harshil@hoppingturtles·
@vaxryy you should do a coding livestream, I'd watch it and learn a few things
English
4
0
46
1.4K
vaxry
vaxry@vaxryy·
first part is massively cleaning up the config manager, probably the worst part of hyprland code atm, over the years it's gotten a nightmare to maintain this will take a while...
vaxry tweet media
English
15
3
305
14.1K
Harshil
Harshil@hoppingturtles·
@vaxryy which Claude model? agent or web UI?
English
1
0
1
1.2K
vaxry
vaxry@vaxryy·
All the AI talk, so I actually tried, but after 400 thousand tokens the result is pretty bad, I am writing this by hand. It will take days instead of 2 hours but at least it will work properly...
vaxry@vaxryy

say less, claude implement pls

English
16
0
300
18.9K
Harshil
Harshil@hoppingturtles·
@elonmusk I thought macro hard would be making AI Games
English
0
1
0
22
Elon Musk
Elon Musk@elonmusk·
Macrohard or Digital Optimus is a joint xAI-Tesla project, coming as part of Tesla’s investment agreement with xAI. Grok is the master conductor/navigator with deep understanding of the world to direct digital Optimus, which is processing and actioning the past 5 secs of real-time computer screen video and keyboard/mouse actions. Grok is like a much more advanced and sophisticated version of turn-by-turn navigation software. You can think of it as Digital Optimus AI being System 1 (instinctive part of the mind) and Grok being System 2. (thinking part of the mind). This will run very competitively on the super low cost Tesla AI4 ($650) paired with relatively frugal use of the much more expensive xAI Nvidia hardware. And it will be the only real-time smart AI system. This is a big deal. In principle, it is capable of emulating the function of entire companies. That is why the program is called MACROHARD, a funny reference to Microsoft. No other company can yet do this.
English
8.3K
11.5K
79.8K
47.6M
vitalik.eth
vitalik.eth@VitalikButerin·
If you've eaten so much that it's not healthy to eat more and you don't want to eat more, and you still eat more anyway "to finish the food", then you are just using your mouth as a garbage can.
English
1.7K
435
6.9K
811.7K
Harshil
Harshil@hoppingturtles·
@charliermarsh don't worry she's just checking out how bad pip is compared to uv
English
0
0
0
772
Charlie Marsh
Charlie Marsh@charliermarsh·
Not kidding just found out my wife is using pip
English
93
31
2.3K
169.6K
Harshil
Harshil@hoppingturtles·
@yunta_tsai I mean they have pick and place machines now, they don't manually solder it
English
1
0
2
285
Yun-Ta Tsai
Yun-Ta Tsai@yunta_tsai·
I asked my kid to do child labor to make her own toy with surface-mount devices. I held the soldering tip, but she had to put that tiny little thing in the center with tiny tweezers and her tiny hands while laying down rosin core. Her reaction was gold — “Did someone actually have to put these things for my toy? This is so hard.” “Oh yes, some people in Asia had to do this every day on an 8+ hour shift.” The soldering didn't look pretty but was functionally okay. We weren't aiming for perfection but for the experience. We went through each soldering joint and pointed out what was good or bad. To be continued…
Yun-Ta Tsai tweet mediaYun-Ta Tsai tweet media
English
16
2
342
10.8K
Harshil
Harshil@hoppingturtles·
@vaxryy didn't notice the performance difference at first, but it was very obvious when playing a browser based game. My FPS went from 40 ish to ~60!
English
0
0
1
172
vaxry
vaxry@vaxryy·
oh hey I can answer this :) benchmarking and debugging GPUs is a pain. They operate as black boxes for the most part, and you have to rely on vendor-specific random-ass half-baked software / solutions for seeing what is the problem and by problem we usually mean something internal in the driver that you have to guess what it actually corresponds to from your code. (and before some "expert" suggests things like renderdoc, these don't work for wayland compositors) minor gpu performance improvements (10-40%) came from us optimizing our shaders a bit, courtesy of Ujin. I wanted to do it for a while, just always found myself doing other things. major gpu performance improvements (up to even 3-4x) came from Tom's (and a bit mine) work on opengl oddities. Drivers can behave very differently on different hardware, and especially when weird constraints come into play (iGPUs, laptops, power states, etc). Tom has identified some places where we caused GPU drivers to stall / do unnecessary work under specific conditions with intel-specific debugging tools. On desktops this pretty much wouldn't ever happen (thus, no bonus from this) but on laptops, especially on battery, it would be apparent. We managed to, in some cases hint, in some cases lie, to the drivers to take faster paths. There are also some minor fifo / dmabuf / sync fixes we made to speed up communication. That's about it. CPU code was fine, we're not stupid, hl cpu code is defo made with speed in mind and profiled from time to time, but GPUs are a big PITA as always :)
vaxry tweet media
English
6
5
315
14.2K
Harshil
Harshil@hoppingturtles·
@sama editing tweets is a thing you know?
English
0
0
0
34
Sam Altman
Sam Altman@sama·
Tonight, we reached an agreement with the Department of War to deploy our models in their classified network. In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome. AI safety and wide distribution of benefits are the core of our mission. Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems. The DoW agrees with these principles, reflects them in law and policy, and we put them into our agreement. We also will build technical safeguards to ensure our models behave as they should, which the DoW also wanted. We will deploy FDEs to help with our models and to ensure their safety, we will deploy on cloud networks only. We are asking the DoW to offer these same terms to all AI companies, which in our opinion we think everyone should be willing to accept. We have expressed our strong desire to see things de-escalate away from legal and governmental actions and towards reasonable agreements. We remain committed to serve all of humanity as best we can. The world is a complicated, messy, and sometimes dangerous place.
English
15.9K
4K
34.1K
38.2M