nravic

7 ms·

1.
Ghosts&Angels, Shells, Rhizomes(nravic.substack.com)
2 points by nravic 9d ago|0 comments
2.
Icons as a Service(nravic.com)
55 points by nravic 13d ago|8 comments
3.
by nravic 1y ago
I think even easier in fact - what's happening behind the scenes w/ an LLM is far more opaque
4.
by nravic 1y ago
This is super interesting! We do something similar I think by taking a checkpoint after model initialization. I'm curious what you think about our approach, here's some benchmarks: https://docs.cedana.ai/articles&#
5.
by nravic 2y ago
How were you handling GPU state w/ pytorch? We added some custom code around CRIU to enable GPU checkpointing fwiw: https://docs.cedana.ai/setup/gpu-checkpointing/
6.
by nravic 2y ago
We do this using CRIU right now! https://github.com/cedana/cedana In fact one of our customer's use cases is exactly what you describe, allowing users to "hibernate" container workspaces.
7.
by nravic 2y ago
I'm probably the cofounder of the guy you spoke with! Here's our repo: https://github.com/cedana/cedana
8.
by nravic 3y ago
what kind of pannier bags do you use to haul groceries? All the ones I've used have been too small for my purposes.
9.
by nravic 3y ago
IIRC (and in my experience) KL divergence doesn't account for double counting. Wrote a paper where I ended up having to use a custom metric instead: https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=4...
10.
by nravic 3y ago
This is so cool! Out of curiosity, could you walk through your decisionmaking around using Nix and how it fits in with Rivet?
11.
by nravic 3y ago
how often are these batch jobs run? I'm curious to know what the absolute maximum sync frequency can be.
12.
by nravic 3y ago
Self plug: run llama.cpp as an inference server on a spot instance anywhere: https://cedana.readthedocs.io/en/latest/examples.html#runnin...
13.
by nravic 4y ago
GrapheneOS uses WebUSB to flash your phone. Pretty wild first experience doing it for me - I'm used to the hassle of sideloading a bootloader and then flashing.
14.
by nravic 4y ago
Hey! I'd love to learn more about the position - the hook @ jpl email on your website is bouncing though. Do you mind sending me contact details so we can chat? Thanks :)
15.
by nravic 5y ago
Do people out of college even do Leetcode anymore? Surely there's a better use of time once you're in a paying job.
16.
by nravic 6y ago
If you're looking at aerospace/defense startups, this is usually the case. It's not unheard of for them to employ people who's primary role is grant writing to try and get (for example) SBIR funding
17.
by nravic 7y ago
Zotero is my go-to
18.
by nravic 7y ago
Julia's REPL is beautiful
19.
by nravic 7y ago
Yeah, spiking neural networks are a tough nut to crack. Check out Nengo if you're interested in learning more
20.
by nravic 7y ago
What are Protonmail's problems?
21.
by nravic 7y ago
very visceral reaction there
22.
by nravic 7y ago
I've followed the same path, out of curiosity do you mind if I PM you asking a few questions? I'm a couple months out of undergrad and I'd like to pick your brain a bit.
23.
by nravic 7y ago
This is really cool, what are you training?
24.
by nravic 7y ago
The population argument is fundamentally flawed. It places the burden of climate change onto the global south, which is far from the actual truth
25.
by nravic 7y ago
yes
26.
by nravic 7y ago
pleasantly surprised to see julia used in this
27.
by nravic 7y ago
There's a bunch of Eigen analogues in Rust, which is slightly frustrating. ndarray is pretty great though
28.
by nravic 8y ago
looks like that's a "fuck you"
29.
by nravic 8y ago
Hello, I emailed you a while ago and I got a response back regarding a phone call; however nothing after that. Would you like me to follow up on that?
30.
by nravic 8y ago
I sent you an email two months ago, and asked about a follow up last month (after which I sent an email). So here's another follow up (with which I'll also send another email) in the hopes that you see it. Would love to talk to yo

More ›