HackerTrans
TopNewTrendsCommentsPastAskShowJobs

martianlantern

no profile record

Submissions

Gumbel Max trick for LLM sampling

darshanmakwana412.github.io
3 points·by martianlantern·vor 4 Monaten·0 comments

Strassen's Matmul with AVX 512

martianlantern.github.io
1 points·by martianlantern·vor 5 Monaten·0 comments

System Design Interview: An insider's guide (Alex Xu) [pdf]

bytes.usc.edu
1 points·by martianlantern·vor 6 Monaten·0 comments

3 Months of Using Neovim

martianlantern.github.io
5 points·by martianlantern·vor 6 Monaten·1 comments

Fail Faster on Your Idea

martianlantern.github.io
2 points·by martianlantern·vor 6 Monaten·1 comments

Fail Faster on Your Ideas

martianlantern.github.io
1 points·by martianlantern·vor 7 Monaten·0 comments

Distilling persona vectors into LLM weights

martianlantern.github.io
1 points·by martianlantern·vor 7 Monaten·0 comments

Fast Median Filter over arbitrary datatypes

martianlantern.github.io
38 points·by martianlantern·vor 7 Monaten·2 comments

Updating My Bash Prompt

martianlantern.github.io
2 points·by martianlantern·vor 7 Monaten·3 comments

How to Establish SSH over Tor

martianlantern.github.io
2 points·by martianlantern·vor 7 Monaten·0 comments

Median Filter over Arbitrary Datatypes

martianlantern.github.io
1 points·by martianlantern·vor 7 Monaten·0 comments

Show HN: Python Asyncio Puzzles

github.com
1 points·by martianlantern·vor 9 Monaten·0 comments

LLM Routing Strategies

martianlantern.github.io
2 points·by martianlantern·vor 10 Monaten·0 comments

Paged Attention Performance Analysis

martianlantern.github.io
1 points·by martianlantern·vor 10 Monaten·0 comments

Paged Attention Performance Analysis

martianlantern.github.io
3 points·by martianlantern·vor 10 Monaten·0 comments

comments

martianlantern
·vor 2 Monaten·discuss
This is not something to taunt about
martianlantern
·vor 3 Monaten·discuss
Cool project, but just a side thought I was having about how do people have resources and the money to make things like this and make it avl for public, I mean it's fair to say they have their own GPUs or if they are using api keys for gpt or Gemini with enterprise subsidized inference

But still coming from a frugal background I still cannot wrap my head around this
martianlantern
·vor 3 Monaten·discuss
Very nice! I found the tutorial on gaussian splatting to be very neatly written and educational
martianlantern
·vor 3 Monaten·discuss
Why are they doing this?
martianlantern
·vor 5 Monaten·discuss
Nice! I need something similar for english now
martianlantern
·vor 5 Monaten·discuss
I am awaiting for a full duplex model from qwen team
martianlantern
·vor 6 Monaten·discuss
This is cool! I am more interested in how you guys generated next edit training data from repos, seems like there are lots of caveats here. Would love your insights

Again amazing work! waiting for what you guys cook next
martianlantern
·vor 6 Monaten·discuss
my site: https://martianlantern.github.io/
martianlantern
·vor 6 Monaten·discuss
I liked the new approach but I don't like the pseudo security framing. I don't see how an LLM rewritten query is secure than just sending my raw query, in both cases I don't feel secure at all
martianlantern
·vor 6 Monaten·discuss
Cool! but it would be more educational if it explained each step and it's reasoning
martianlantern
·vor 6 Monaten·discuss
If I am missing something, isn't this the whole purpose of hacker news to be a link aggregator for technical/developers?
martianlantern
·vor 6 Monaten·discuss
I have given up on any external bash configurator a long time ago, instead I write my own bash prompts these days, they lack a lot of functionalities but I am much happy with them for now, also a shameless plug: https://martianlantern.github.io/2025/11/updating-my-bash-pr...
martianlantern
·vor 7 Monaten·discuss
would be great if there was a chat room where we can also chat with all the other people in the lobby :)
martianlantern
·vor 7 Monaten·discuss
damn
martianlantern
·vor 7 Monaten·discuss
Wow this seems very interesting! Can we get a TLDR of how this was achieved?
martianlantern
·vor 9 Monaten·discuss
This is awesome! It reminds me of this https://github.com/lucidrains/self-reasoning-tokens-pytorch
martianlantern
·vor 9 Monaten·discuss
The article itself seems to be AI slop, too many em dashes ...
martianlantern
·vor 10 Monaten·discuss
Do mean hard to be used because of compute required to render it or because of setting up and understanding the training pipeline and related hyperparameters?