just yapping about whatever.
well.
I wanted to blog about Agentic Engineering but not really in the mood for serious stuff right now
]]>alright I do want to visit Japan/Korea late this year but that would have to be in like October at least, and it’s only May now
apart from small trips, say I leave for Australia (?) in early August? that’d be 2-ish months to October (maybe I just do Japan), then…back to Australia? back to Singapore? something else?
like:
3 months in Australia into 3 months in Japan/Korea?
]]>guess what I’m writing concurrent code now and it’s super fucking mindblowing wtf this is Hard
Ugh
…
would you rather be cringe in the pursuit of being kind, or cold in the pursuit of being cool?
]]>I saw but didn’t say anything
the clang of his collision was something to behold…he wasn’t hurt, but I died laughing, and even now thinking of it brings a smile to my face (I swear we’re good friends)
‘I pass the test,’ she said. ‘I will diminish, and go into the West, and remain Galadriel.’
this is one of the most powerful scenes (IMO) in the whole collection
give me strength on my quest to give up my ring
]]>but I realised I don’t wanna do that because I don’t want to know who’s reading, because ultimately I am writing for myself
do you ever get one of those moments where
okay so there was this girl whom I shared one class with in JC (rare subject) and we were like…not super close? from my perspective at least
once I pretended I broke my leg and she got super anxious (and then mad when she realised it was a joke)
then one day she sent me an email with a pic that was literally a BUNCH of hearts and that said “love, from your <cute way to spell her name>”
it was like 6-7 years later that I looked back and was like “oh wups”
can’t say I was ever interested in her but it’s one of those situations where I never really considered it, right…because I know whom I was crushing on (LIKE MAD BTW) at that time…………who knows, right
after I thought of her (the cute email girl, not my crush) I went to stalk her; she worked as a teacher for a few years, and now she’s doing product at an EdTech company (!)
entirely seriously (no false self-deprecation), of everyone I have liked and everyone who has liked me, she might be the one whom I am least worthy of
like she’s not making a ton of $$, and she’s not super smart (smart enough), nor is she super good-looking (but defo pretty if you consider it objectively), but she is really a good person, and I can say with certainty she is a better person than I am
life is really strange. I remember, a long time ago, I was like…6? and riding on a kick scooter with my father
so we fell down and both got injured, and like a few days later he commented on how I healed super fast and he didn’t any more. well, he was a bit older then than I am now, but…sometimes I still feel like me that day I got that cute email. clueless, drifting through life, not really sure what the best path is.
being a manager is actually so fucking hard and idk how I keep it together
the crazy thing is that my underlings say weird ass stuff like “you are the best boss I have ever had” and I’m like ?? yo I’m just trying to make a living here this is like basic
nobody sees the insane anxiety I have when I need to have The Talk about performance, and scolding people is hard. not micromanaging is hard (when someone on top is projecting their anxiety on you, it takes all I have to not then project my anxiety on my underlings). firing is hard, so gut-wrenchingly hard, like wdym I have to tell this person “yo this is your last day here”???? like I’m really not cut out for this and I prefer just being an IC because I can do what I want and doesn’t afraid of anything
how about we just start driving and go a long, long way? we can park in the middle of the desert and kiss beneath the stars, and help each other forget. and maybe in time, your green eyes will turn blue, and my collarbones will harden again.
]]>imagine this
everyone thinks you are super in love. you leave the house at 7:30 AM to buy something for your girlfriend to take to work.
on the way back you start thinking about how this might literally be the last time you do this, and start tearing up, but you don’t want to cry in a public place.
you get back and start the painstaking task of separating the meat from the bones so she doesn’t have to do it later. you are alternately bawling and sobbing all the while.
some time later she wakes up and sees you (still peeling, it takes a while) obviously distraught.
“what’s wrong, babe?”
what’s wrong? how can I even begin to explain what is wrong? what words can I find to describe how the shards of our love chipped away, bit by bit, and fell forever into the long silence that is the only thing that will remain when we walk in opposite directions for the final time?
where else can you find two hearts with us-shaped holes in them? nobody ever falls in love thinking that they will be ground down by the emotionless boot of reality, sheer as a cliff face, featureless as Slenderman. nobody ever thinks, in all the sublime hubris of youth, that they will be unable to buck the odds. they think that love conquers all, and that is true, but not for the reason they think.
“love conquers all” really means that obsession and devotion are two sides of the same coin, and seeing it won’t work with your own eyes is no panacea for the bottomless ache inside. it means that love takes your resilience, your restraint, your self-preservation instinct, and your good sense, and whomps them a good one. it pounds them until they don’t know left from right, and hangs them out to dry.
well.
so anyways…I want to talk about Subquadratic.
as a brief primer, in computer science, we use the term “time complexity” to talk about how long certain tasks take on average.
for example, finding a given card in a deck of standard poker cards is done in “linear time”, meaning that if the number of cards doubles, the time taken also doubles (because you need to go through them one by one, but you only need to traverse the whole deck once)
now, imagine you have a box of socks. because you are a neat and orderly person, you want to sort them into pairs. however, because you are cursed by the Genie of Analogies, you can take out only two random socks at a time; if they match, you can set them aside. if they don’t, you have to put them back.
on average, if the number of socks in the box doubles, you’ll take four times as much time to sort them; we say this is done in “quadratic time” (because for each sock, you need to go through the entire box once, so the number of searches you do is proportional to the number of socks squared).
the fundamental mechanism of LLMs is called “attention”, where, for every token (group of characters) in the input, the LLM calculates the relevance of every other token. this maps nicely on the sock analogy (if you don’t look too hard): every sock needs to be considered in relation to every other sock, and every token needs to be considered in relation to every other token. the attention mechanism therefore also takes quadratic time.
when I read the original paper, I was like “yo this is insane”, because in general, quadratic algorithms are only practical for small inputs; definitely not whole ass books. but thanks to Moore’s law, we made it work!
however, we’re at the point where scaling further is impractical; despite the frenetic pace at which new data centres are being constructed, the industry has settled around “context management” - various techniques to limit LLM input, because otherwise it’s just too expensive and slow (and the quality of the output deteriorates rapidly anyway).
enter Subquadratic’s SubQ model, which claims to be a subquadratic (i.e. linear) LLM. that in alone isn’t groundbreaking; many have done that before. no, their real breakthrough, they claim, comes in doing that while matching or beating state of the art models.
if their claims are indeed valid, this would be a pretty big shift. it would mean the ability to ingest and reason across whole codebases, books, and other such repositories that are currently Way Too Big, opening up whole new avenues of LLM use.
deciding whether that is a good or bad thing is left as an exercise for the reader. the ethics of LLM usage and operationalisation (including the consumption of water and other natural resources) aside, though…the science is pretty cool.
]]>