Viewing profile — fallmonkey

Appreciated the response! I noticed the same when I ran tau2 myself on gpt-5 and 4.1, where gpt-5 is really good at looking at tool results and interleaving those with thinking, wh…

Correct, and with reasoning, the ratio is totally off. As others have pointed out, actual usage is way higher (much more than 3-5x) than the estimation in the article, which is pro…

The estimation for output token is too low since one reasoning-enabled response can burn through thousands of output tokens. Also low for input tokens since in actual use there're …

While there're interesting findings here, https://arxiv.org/pdf/2502.03373 (also with a lot of good findings) suggested some contradicting theory on the critical mass of training p…

Strangely, deepseek has been always a prominent name in open source LLM community since last year, with their repos and papers - https://github.com/deepseek-ai . Nothing of it is r…

There’s an interesting dynamic here. What value of OpenAI to use for conversion calculation? 86b round is pretty much dead if they move to MS, yet 29b is too low (even 86b is low i…

Over the years, I've experienced this many times. After being addicted to something (game/movie/YouTube/etc.) for lengthy hours, mental refractory period kicked in so I enjoyed a l…

I'd recommend Julie Zhuo's The Making of a Manager. She's got a twitter https://twitter.com/joulee where you could check out her sharing of managerial wisdom to have your own gauge…

On personal anecdotes, I've used the same internet identity across all platforms and it's not hard to locate my real life identity from there...since like 20 years ago when I got o…

"A common question I get is about work/life balance. And I view it a little different too now. At Facebook I could get by barely working a few hours a day. But the job was so unsat…

Yet you might be surprised to know that, in China, all those "offline games" are categorized as online games because they all can be distributed online, when it comes to import rev…

With Google doing front loading(33/33/22/11) and other companies also shifting away from conventional 25% per year vesting to make first few years more attractive, they gotta catch…

I'd like to provide an personal take on this - this will be a bad deal until Uber can be a super app doing 100x more things than it's doing today. So maybe never. Why? Because exac…

The title misses the other, and more impactful/unprecedented change of giving everyone "exceeds" rating and skipping perf review for them (except for those high performers who will…

But China is making rather good progress so far so maybe something is able to stop the spread?

I’ve also read this book a couple times but sadly as a engineer in large and mature company I hardly find any opportunities to actually apply or strengthen the learning, then as ti…

This is indeed very interesting to see and made me want to share my stories. I've also scanned over thousands of books (mostly Japanese manga, in a way to digitalize them and make …

28% know what it is? That's more surprising to me actually.

My failed attempt to find ANY information about his father somewhat proves your theory, or google doesn't wanna tell me?

I've lived in different cities in China and US. 4-6 RMB to 1 USD in major cities like Beijing/Shanghai, and 1-3 RMB to 1 USD in less developed cities or sururban area. Less develop…

Hey David, thx for the kind advice. I did ask my current manager about this decision and he did mention that he wants to find a engineering project manager (we have PMs) who could …

Exactly! I've read many HN stories about bad interview experience and I really wish to make it an enjoyable one for both me and candidate by staying away from those new-grad leetco…