Eugeleo
Eugeleo on Hacker News
Posted by
Eugeleo on Wednesday October 22, 2025 @08:08PM
from the ask-hn dept.
Writing research code (in my case ML/AI) is very different to writing production code. The goals are different, and thus so are the best practices, patterns, and values.What's your favorite resource on how to write code in research? What are the research-code-specific equivalents of Rich Hickey's talks or SPJ's posts or the many many SWE blogposts posted to HN?
FWIW I think this is exactly the opposite of what the OP was trying to say. I think the OP meant that they often wish authors had used more space to explain their ideas, because that would immediately make them more accessible.
And I must agree. I often feel papers are written with an extra 20% over the page budget, then condensed in the last minute to fit the constraints, which hurts the exposition.
Posted by
Eugeleo on Thursday August 28, 2025 @07:52AM
from the ia.net dept.
Posted by
Eugeleo on Tuesday August 19, 2025 @12:03PM
from the twitter.com dept.
Posted by
Eugeleo on Monday June 23, 2025 @06:43PM
from the politico.com dept.
Posted by
Eugeleo on Saturday December 07, 2024 @01:38PM
from the arxiv.org dept.
Everyone in the comments swems to be in love with KM. I wonder, what do you all use it for? It sounds like I’m missing out on a lot, but when I’m thinking about it, nothing in need of automation comes to mind. That is, except for opening a few common apps and window management. What am I missing?
Great! Any up-to-date guide to get the latest PyTorch running on Macs? Do we still have to use conda for example?
Hey, mind sharing your workflow? Where do you read papers? And do you annotate them?
Posted by
Eugeleo on Wednesday April 20, 2022 @03:27AM
from the hogbaysoftware.netlify.app dept.
Posted by
Eugeleo on Monday December 20, 2021 @04:10PM
from the medium.com dept.
I mean, it only messes with the efforts of the advertisers, right? Not that they can detect how many views the sponsored block has. Ofc if this becomes too popular, the ad industry will eventually catch up, but this is long ways away imho.
I've never used nix, but it sounds interesting. What getting started guide would you recommend? Ideally one that has all these "insider tips" that you offer, so that I learn to use Nix "properly".
Thanks for the summary, it is very helpful! I love Haskell and its philosophy and I'm intrigued to find that there can be some of it in a package manager as well.
One question: How do you properly install things like R on the M1 Macs, then? Do you use a 'backup' package manager?
Hey, a bit off-topic, but what do you think about only buying a 16GB version? According to a YT video I saw, the 16GB and 32GB Macs are roughly equal in terms of performance. [1]
I'm on the fence between buying a 32GB 24 core GPU Max version, or going for a cheaper 16GB 14 core GPU and using the leftover money to buy a home server with good GPU and RAM.
I'd like the Mac to sustain 95% of my programming, design, and data science needs, but I'd be willing to run the last 5% (gaming, very deep neural networks, or big datasets) remotely. Especially if I could get more bang for my buck buying a home station instead of maxing out the MacBook.
[1]: https://www.youtube.com/watch?v=gqiYSt4nAFs
Thanks for the viewpoint l. I've been wholly satisfied with my old MacBook, though, and thus don't think the overhead of switching systems is worth it for me.
Just out of curiosity, which Dell laptop would you recommend? Ideally a 14", light, and with good battery life. (I'll leave the more technical specs up to you)
Apple neural engine officially supports Tensorflow, I'm not sure if the GPUs are good enough to bother with it, though.
Posted by
Eugeleo on Tuesday October 19, 2021 @02:45PM
from the ask-hn dept.
I've read through the MacBook threads and seen one question repeated again and again: which specs to go for in the new MacBooks. I figured it might be worth it to make a dedicated post to discuss this.To start this off, consider the following scenario:- 50% of my work is web or mobile app development- another 50% is training ML and deep learning models
(currently on a cluster, but I'd prefer...
If you find out, let me know! I'd love to know this, too.
Thanks. I don’t really like working at home, though I do plan to a have a nice external screen there (just can’t decide between high density and ultrawide). Maybe I should just ditch the caffes at get used to my desk.
Posted by
Eugeleo on Saturday October 16, 2021 @03:54AM
from the ask-hn dept.
With the new MacBooks around the corner, I'm torn between the 13" and 15" (or more likely, 14" and 16") versions. All things besides size and weight equal, what notebook screen size do you prefer and why? Doesn't have to be MacBooks of course, I'm interested in all thoughts on this topic, especially from people who tried both.I can't decide whether being on the move often is an argument for the...
Posted by
Eugeleo on Tuesday October 12, 2021 @04:48AM
from the github.com dept.
Posted by
Eugeleo on Tuesday October 12, 2021 @04:35AM
from the self-post dept.
(forgot to include the url, please delete this)
I also couldn’t find the pricing info, so I just blindly signed up for the closed beta. I’ve been guided through a short form, at the end of which I’ve learned that after a 14 day trial Sunsama will cost $20 a month (!), even for personal or any other non-commercial use.
This is what I use it for! I got inspired by the book Make Time; I always have a lot of things on my plate, and I have a tendency to overfill my todo list with tasks which ultimately leads to me feeling bad for not managing to complete everything I set out to.
I'll check them out, thanks. Sorted is nice, but not on Things-level of polish.
See my other comment in this thread.
AFAIK, no, unfortunately. But maybe if enough people request it...
A lot of people are asking this! Clearly they could do some work on marketing and communication. Let me quote my other comment in this thread:
Basically, it is a todo manager, but with built-in first-class scheduling. You input your todos, assign time estimates, and Sorted helps you spread them out in your day, keeping your calendar events in mind.
To explain the value in this, let me copy one of my comments in this thread:
[...] Sorted is really great for people like myself who tend to overfill their todo lists, and subsequently feel bad at the end of the day when they don't manage to complete everything.
Sorted pushes me to assign time estimates to the task, which helps me to keep my todo-list for the day more realistic and manageable. I like it more than managing my todos directly in the calendar [...].