I'll probably relatively off social media as I navigate a big move & new job, but I would love to in-person connect with folks living in NY working on cool things! Ping me if you're in town.
09.02.2026 22:16 β π 6 π 0 π¬ 0 π 0@alphaconvert.bsky.social
cutler.pl
I'll probably relatively off social media as I navigate a big move & new job, but I would love to in-person connect with folks living in NY working on cool things! Ping me if you're in town.
09.02.2026 22:16 β π 6 π 0 π¬ 0 π 0Seems like a good time as any to say I'm:
(1) moving to New York next week, where I'll:
(2) be writing OxCaml at Jane Street!
Excited to start the next thing :)
Thanks, Anil!!
09.02.2026 21:59 β π 0 π 0 π¬ 0 π 0Thanks, Rachit!
09.02.2026 21:59 β π 0 π 0 π¬ 0 π 0Thanks David!
09.02.2026 21:57 β π 1 π 0 π¬ 0 π 0Successfully defended my dissertation today. Now Dr. Joe!
09.02.2026 17:55 β π 49 π 1 π¬ 12 π 1Still OCaml
30.01.2026 20:27 β π 14 π 2 π¬ 1 π 0this is the one posting the AI-"upscaled" pictue?
26.01.2026 17:19 β π 13 π 0 π¬ 0 π 0Turns out it's a tiny fix. Not entirely sure how to test it. π€·ββοΈ
github.com/alpha-conver...
maybe I should just write the PR that fixes this
25.01.2026 18:49 β π 1 π 0 π¬ 1 π 0After some investigation, it seems like the oxcaml folks are already well aware of this
@thenumb.at :)
github.com/oxcaml/oxcam...
seems like this bit of the compiler is still in flux, and there are some sharp-ish corners. I ran into this codegen bug with unboxed bigstring accesses a few weeks ago while working on 1brc... you can cause a segfault with this if you try really hard
gist.github.com/alpha-conver...
many of these do exist, but seemingly not all of them...
github.com/oxcaml/oxcam...
hm is there no caml_bigstring_get..._indexed_by_int16#?
25.01.2026 18:32 β π 0 π 0 π¬ 2 π 0oops
25.01.2026 18:12 β π 0 π 0 π¬ 0 π 0omg yay toplevel unboxed bindings!
25.01.2026 17:59 β π 1 π 0 π¬ 1 π 0Nonviolence is, in fact, working. This administration is weaker than it was a year ago. More and more of the public is becoming galvanized against it; its agents are being impeded.
24.01.2026 21:53 β π 8907 π 2098 π¬ 96 π 63Wow, exciting!!
22.01.2026 01:14 β π 0 π 0 π¬ 0 π 0this claude was the unofficial mascot of the other claude
21.01.2026 22:15 β π 2 π 0 π¬ 0 π 0Hey, I'm looking for speakers at DC Systems for this year! Let me know if you, or anyone you know would enjoy giving a talk to an extremely nice and curious audience!
15.01.2026 01:44 β π 2 π 1 π¬ 0 π 0Itβs called βvs codeβ because the code is an enemy you are fighting
15.01.2026 00:54 β π 753 π 181 π¬ 12 π 4this fucking rocks
13.01.2026 19:12 β π 1 π 0 π¬ 0 π 0This one really bugs me
05.01.2026 15:24 β π 1 π 0 π¬ 0 π 0also, Applicative usually being defined in terms of (<*>) and not liftA2 was so incredibly damaging.
05.01.2026 14:37 β π 25 π 4 π¬ 8 π 0unfortunately not
05.01.2026 13:24 β π 2 π 0 π¬ 1 π 0try the audiobook itβs so much worse
05.01.2026 13:22 β π 2 π 0 π¬ 1 π 0wow this *is* the most embarrassing possible thing
05.01.2026 13:17 β π 2 π 0 π¬ 0 π 0It now has a feedback category! I wrote a long note in there β hopefully it gets to somebody useful.
17.12.2025 17:54 β π 1 π 0 π¬ 0 π 0Ok I'm really excited you're blogging, but also I'm going to miss the "What should I put here? What were they expecting?" page
08.12.2025 20:26 β π 0 π 0 π¬ 1 π 0Theyβre killing people as a twitter joke. Itβs depraved. Canβt let yourself lose the capacity to be appalled and disgusted and outraged by this.
05.12.2025 02:28 β π 13763 π 4246 π¬ 446 π 176