aportnoy

4 ms·

1.
by aportnoy 7mo ago
Is Bel a smaller LISP than what's described in The Roots of Lisp? [1] [1] https://paulgraham.com/rootsoflisp.html
2.
Train an RL agent to play tic-tac-toe(github.com)
3 points by aportnoy 8mo ago|1 comments
3.
by aportnoy 8mo ago
aportnoy.com
4.
Show HN: Reinforcement learning tic-tac-toe in C, annotated(github.com)
2 points by aportnoy 8mo ago|0 comments
5.
Pretraining Large Language Models with NVFP4(arxiv.org)
1 points by aportnoy 1y ago|0 comments
6.
by aportnoy 2y ago
For NVIDIA, 1. play around with the NVPTX LLVM backend and/or try compiling CUDA with Clang, 2. get familiar with the PTX ISA, 3. play around with ptxas + nvdisasm.
7.
by aportnoy 3y ago
Well then… test your memory :)
8.
by aportnoy 3y ago
Don't overclock your memory.
9.
by aportnoy 3y ago
Wow I came back to post this exact reply. I set my system to a slightly high frequency, ran memtest overnight with errors. Set it back down to a supported frequency, ran a full memtest suite again with no errors. Never had any issues since.
10.
by aportnoy 3y ago
> But the lack of ECC was a huge bummer at the time of purchasing my system. Why?..
11.
by aportnoy 3y ago
> extremely opinionated I have not seen a single codebase that widely uses uint8_t and does not typedef it to u8. It is the exact opposite of "extremely opinionated".
12.
by aportnoy 3y ago
Note Sam Altman is a Cerebras investor.
13.
by aportnoy 3y ago
Asking because I am on that team :) I went through the same hiring process and had a positive experience at every stage. I had a strong competing offer but went with the JAX team at NVIDIA. I'll pass it along as feedback.
14.
by aportnoy 3y ago
What happened?
15.
by aportnoy 3y ago
1 is the multiplicative identity 0 is the additive identity all([]) is True any([]) is False
16.
by aportnoy 3y ago
Go to the blog and skip to results: https://ai.meta.com/blog/seamless-m4t/
17.
Visualize core-to-core latency(github.com)
2 points by aportnoy 3y ago|0 comments
18.
by aportnoy 3y ago
The project is a port of https://github.com/nviennot/core-to-core-latency from Rust to C.
19.
Show HN: Visualize core-to-core latency in ~200 lines of C and Python(github.com)
2 points by aportnoy 3y ago|1 comments
20.
by aportnoy 3y ago
I used to surf near Scripps Pier while in college and I remember there were always a couple people sitting outside in beach chairs working on their laptops. Salk Institute is an even more surreal place.
21.
by aportnoy 3y ago
The funny thing is Perl is now arguably more obsolete than sed and awk.
22.
by aportnoy 3y ago
I recently interviewed with a fantastic startup in that space called TestFit (testfit.io).
23.
by aportnoy 3y ago
This definitely isn't "printing" in the original sense, just a serialization algorithm.
24.
by aportnoy 3y ago
In the original LISP with dynamic scoping I think it would be fair to serialize that expression as is: "just use whatever value is bound to 'y' at the time of evaluation, if any".
25.
by aportnoy 3y ago
Everything is an object that is either a cons cell or an atom. Associate the address of each object with a unique index. Serialize each object as a pair (index, obj), where 'index' maps back to the original address where the objec
26.
by aportnoy 3y ago
Yeah, something like "a" in the below code wouldn't have a "natural" serialization I don't think: (define a (list "a")) (set-cdr! a a)
27.
by aportnoy 3y ago
> In general lisps _can't_ print arbitrary objects in a form that can be read back in. Hmm, why not?
28.
by aportnoy 3y ago
I mean, his simulator already has a disassembler contained within it, would just need to replace comments with print statements.
29.
by aportnoy 3y ago
How about a RISC-V disassembler in 200 lines of C99? https://github.com/andportnoy/riscv-disassembler/blob/master...
30.
Apache log pretty printer in 130 lines of C(github.com)
2 points by aportnoy 3y ago|0 comments

More ›