bartwr

4 ms·

1.
by bartwr 5mo ago
I have most likely outdated info, I left Google Research 4y ago. Back then, available TPU instances were plenty and GPU scarce. Nobody wanted to mess with an immature crashing compiler and very steep performance cliffs (performance was exce
2.
by bartwr 5mo ago
My experience cycling regularly in NYC: bike lanes separated by curb, stoppers, or poles are more dangerous as cars stop at their entrances/exits and I am literally trapped or cannot enter them before/after an intersection. I&#x27
3.
by bartwr 5mo ago
I love using AI tools and they are changing my work and life in amazing ways. I cannot imagine going back. And yet, I am more concerned about the social damage due to their widespread use and the amounts of slop they generate. Just this we
4.
by bartwr 7mo ago
It's hilarious that people describe their anecdotal experience of being in a calorie deficit as a proof that "intermittent fasting works". This was never the question, but whether intermittent fasting brings additional weight
5.
by bartwr 2y ago
I share the sentiment. I haven’t used it in a while (at work use different languages and in the last few years my personal coding is only Python script/Jupyter notebook bite-sized), but anytime I hop into it, it immediately "click
6.
by bartwr 2y ago
Original author here and it's been a while since I have read such word salad nonsense, sorry. Why people who have no idea or expertise comment on articles? GenerateMips API constructs a mip chain by using box/bilinear (equivalent
7.
by bartwr 2y ago
It wasn't an exaggeration! :) The shock of many of my colleagues (often not even junior... sometimes professors who decided to join the industry) "wait, I need to talk to product teams and ask them about their needs, requirement
8.
by bartwr 2y ago
You don't understand the Silicon Valley grind mindset :) I personally agree with you - I am happy working on interesting stuff, getting a good salary, and don't need a promo. Most times I switched jobs it was a temporary lowering
9.
by bartwr 2y ago
If someone believed they will earn 2-5x better than in academia, with full freedom to work on whatever interests them, and no need to deliver value to the employer... Well, let's say "ok", we have all been young and naive, bu
10.
by bartwr 2y ago
One unique and new feature of Slang that sets it apart from existing shading languages is support for differentiation and gradient computation/propagation - while still cross-compiling generated forward and backward passes to other, pl
11.
by bartwr 2y ago
It does! Both platform-specific compute shaders as well as cross-compilation to CUDA. The authors even provide some basic PyTorch bindings to help use existing shader code for gradient computation and backpropagation in ML and differentiabl
12.
by bartwr 2y ago
It's interestingly disingenuous that many claim of GLP-1 agonist miraculous effects on all kinds of health problems, where the same problems are "simply" solved by getting on a calorie deficit and lean. Liver, kidneys, heart,
13.
by bartwr 2y ago
Yes, I am an (amateur) photographer for the last 27 years, from film, DSLRs, mirror less, mobile. And I worked on camera ISPs - both hardware modules, saving RAW files on mobile for Google Pixel, as well as software processing of RAWs. But
14.
by bartwr 2y ago
It's not. a) compression can be lossless. b) RAW is not about storing literal photons ADC measurements. It always has "some" processing as those always go through an ISP. We can obviously discuss which processing is the cutof
15.
by bartwr 2y ago
It's the other way around - in hearing, phase is almost irrelevant. At medium frequencies, moving head by a few centimeters changes phase wand phase relationships of all frequencies - and we don't perceive it at all! Most audio sy
16.
by bartwr 2y ago
Seems you have not worked with ML workloads, but base your comment on "internet wisdom", or worse, business analysts (I am sorry if that's inaccurate). On GPUs, ML "just works" (inference and training) and are alway
17.
by bartwr 3y ago
Yes, my experience with academics is that there are a lot of very dishonest people. They are political bullies who also lie in their research. Chances of being caught are close to zero (I have contacted many times authors of papers who&#x27
18.
by bartwr 3y ago
I lived in SF for ~1.5y, and it's not NYC, and I did not like it too much, but it certainly has some of the city conveniences and is not a car-hell suburb. (I lived in a building with ~10 units in Castro, which was cool) But my questio
19.
by bartwr 3y ago
I know this is an unpopular opinion in the US, but the tenements can be pretty great. I grew up in Eastern Europe (Warsaw) in "commie" blocks and there was a lot of valid criticisms and problems (like poor quality of buildings, sm
20.
by bartwr 3y ago
I enjoyed the post and appreciate the author sharing their perspective. It's one of many valuable datapoints for anyone considering such a transition. I agree with a lot, but like others - disagree with some. My background: I've w
21.
by bartwr 3y ago
I'm a former game dev and I used ImGui a lot and don't think it's used because those reasons. It's used for quickly hacked debug tools to interleave UI and regular logic and not do a logic/view separation (as it wou
22.
by bartwr 3y ago
Wait, what? This is literally not the point of research. The point of research is to form some theory and gather data that is relevant, and then analyze statistically. Plus propose follow up theories and studies. This is how almost every re
23.
by bartwr 3y ago
I'm always for people over financial entities and capital holders, however in this case it's pretty simple - you sell control to someone for a ton of money, so why on Earth would you expect to retain it forever? You literally got
24.
by bartwr 3y ago
I shipped a dozen products with them (mostly video games), so there's nothing "wrong" that would make them unusable. But programming them and setting up the graphics pipe (and all the passes, structured buffers, compiling, bi
25.
by bartwr 3y ago
If they create a better tool chain, ecosystem, and programming experience than CUDA and compatible with all computational platforms at their peak performance - awesome! Everyone wins! Until then, it's a bit funny claim, especially cons
26.
by bartwr 3y ago
How do I know they didn't use them? They already did something with my biological samples (storing for a different purpose than when they drew my blood) without my consent nor informing me. And also - could eg. police use it?
27.
by bartwr 3y ago
Something super creepy that happened to me recently: a hospital where I've been to a few months ago called me and asked me to participate in some DNA analysis program. They said "oh and the best part? You don't need to do any
28.
by bartwr 3y ago
As an engineer, you generally don't know which one applies. Even qualified lawyers don't decide on the actual criminality (a jury or judge after a valid trial does). So I assume that "reporting wrongdoing" encompasses re
29.
by bartwr 3y ago
"non-profit nature of OpenAI would not allow for it." How so? Seeking more efficiency and cost-effectiveness absolutely does not conflict with a non-profit mission.
30.
by bartwr 3y ago
"Conflict of interest" is not defined by a bad outcome or malice. It is defined as the potential of those, human nature, and various cognitive biases. "Conflict of interest" is disclosing anything that could lead to

More ›