israrkhan

6 ms·

61.
by israrkhan 2y ago
am I the only one who is scared of "it can run any command in your terminal"?
62.
by israrkhan 2y ago
You can export from Lucid to SVG. It also has its own version history. For me it has been working OK.
63.
by israrkhan 2y ago
There are many embedded platforms that do not gcc/llvm based compilers. Also most companies making those platforms are not good at updating their toolchains. Expecting developers to compile their own toolchain, that is unsupported by p
64.
by israrkhan 2y ago
Most important aspect of C is its portability. From small microcontrollers to almost any computing platform. I doubt that any new version of C will see that much adoption. If I want to live on cutting edge I would rather use C++2x or Rust
65.
by israrkhan 2y ago
Here are few reasons that justify its existance. * Different form factor * Not tied to Apple Ecosystem. * Price * You can even use it independently (without phone).
66.
by israrkhan 2y ago
AGI does not need to be based on biological intelligence. it is analogous to human will to fly, and our models were birds, but eventually we came up with something else (airplanes), that are much better at flying than birds (in some regard
67.
by israrkhan 2y ago
You can open-source your FPGA designs for wider collaboration with the community? wider collaboration. Also, FPGA is the starting step to make any modern digital chip.
68.
by israrkhan 2y ago
I was also blown away at CNC grinding step. Perhaps he did it because heat gun would have impacted other parts of the PCB. With grinding the impact was limited to Nand chip only. He eventually had to use heat to cleanup the residue.
69.
by israrkhan 2y ago
You need bazel if you need starlark & sandboxing
70.
by israrkhan 2y ago
Great... a new era of autogenerated podcasts is here.
71.
by israrkhan 2y ago
This is excellent
72.
by israrkhan 2y ago
3.36 Trillion as of today.
73.
by israrkhan 2y ago
I agree with this sentiment. However maybe in few years AI will get better enough that you wont need to debug it?
74.
by israrkhan 2y ago
Roku supports both airplay (mac, iphone, iPad) and miracast (windows and some android devices). Most android devices support Miracast, but Google abandoned support for miracast in their firstparty devices in order to promote their proprieta
75.
by israrkhan 2y ago
The author should have included a Makefile. You need to add -lpthread to command provided in README.md
76.
by israrkhan 2y ago
QEMU is a great piece of software. I use it regularly to debug various cross compiled kernel images under gdb without requiring actual ARM hardware (buildroot + custom kernel + gdb).
77.
by israrkhan 2y ago
This is awesome. In past I worked on adding support for meta-data queries to faiss DB, by hooking up faiss db with SQlite, but this approach (Sqlite extension) is much more cleaner and better. Not sure about performance though.
78.
by israrkhan 3y ago
I am primarily a nevim users. Occasionally I use vs code and dislike it for its slowness. Recently discovered `Zed`[1] on Mac. it is quite fast and really good GUI editor (vs code replacement in some ways). I think NotePadNext will still re
79.
by israrkhan 3y ago
It gets stuck (browser tab becomes unresponsive) when i try to run the default program on main screen. Using chrome
80.
by israrkhan 3y ago
I think you may want to look at OpenAPI specs. It is yaml description of API and can be consumed by many tools.
81.
by israrkhan 3y ago
interesting. This is exactly what I was thinking about. Thanks for sharing
82.
by israrkhan 3y ago
MoE (Mixture of Experts) is an effective way to scale transformers. Gemini 1.5 is already doing upto 1 million tokens. I have not seen any large scale mamba model, so not aware of its shortcomings, but I am sure there are tradeoffs. It shou
83.
by israrkhan 3y ago
When I was at Microsoft we did not use to consider UAC as a security boundary, since there were ways to bypass it. (I used to work on windows kernel security). I wonder how security team let this feature get shipped. Or perhaps now they con
84.
by israrkhan 3y ago
NAT is the main reason slowing down (or preventing) IPV6 adoption.
85.
by israrkhan 3y ago
if you are asking about Apple silicon based Macs, they have integerated GPUs and do not have dedicated graphics memory (UMA) For running 4 bit quantized model, with 70B parameters you will need around 35G Ram to load it in the memory. So I
86.
by israrkhan 3y ago
M1 with 16GB Ram will barely run codellama:13b
87.
by israrkhan 3y ago
This is interesting but nothing new. This idea has been posted on hackaday several times in past. Some even went to stream video 2009: https://hackaday.com/2009/06/05/ds-optical-mouse-camera/ 2014: http
88.
by israrkhan 3y ago
I hope someone will soon post a quantized version that I can run on my macbook pro.
89.
by israrkhan 3y ago
This looks really good..
90.
by israrkhan 3y ago
I tried fauxpilot to make it work on my own llama.cpp instance, but didn't work out of the box. Filed a github issue, but did not get any traction. Eventually gave up on it. This was around 5 months ago. Things might have improved by n

More ›