so true
30.10.2025 13:46 β π 0 π 0 π¬ 0 π 0@sdivvela.bsky.social
Undergraduate at UMD, interested in PL, systems, and security and all things in their intersection :) also avid manga reader, cat lover, and occasional gamer
so true
30.10.2025 13:46 β π 0 π 0 π¬ 0 π 0me: ugh god you know how you make plans for friday night but then friday night gets there and you don't want to do it and it feels sooooooo good to cancel?
my liege lord DaimyΕ Nobunaga (unsheathing katana): even a dog understands loyalty. She whose word is not bond does not deserve to live
thatβs a really beautiful theme
16.10.2025 22:50 β π 5 π 0 π¬ 0 π 0yeah i feel like a lot of βtypes with baggageβ is just some simple case of dependent types
04.10.2025 22:18 β π 3 π 0 π¬ 0 π 0WOOOOOO what is it :0
04.10.2025 01:40 β π 0 π 0 π¬ 0 π 0best of luck!
02.09.2025 21:44 β π 3 π 0 π¬ 1 π 0part of my resentment towards LLMs is to do with them crowding out anything else we could be talking about. there are so many things I could be reading to get better at my craft but I'm presented with "you should actually get worse at it on purpose because nothing matters any more"
08.08.2025 19:27 β π 664 π 130 π¬ 7 π 22kinda wild he was talking about verification at an algorithms and architecture conference though
09.08.2025 04:49 β π 0 π 0 π¬ 0 π 0actually I was at SPAA last week and one of the keynote talks was a professor talking about how LLMs were going to be the future of coding and how it's really important to work on verification for LLM generated code π
09.08.2025 04:49 β π 0 π 0 π¬ 1 π 0Didnβt want to believe this was true. Itβs true.
07.08.2025 21:13 β π 14051 π 3465 π¬ 390 π 311congratulations !!!!
19.06.2025 22:06 β π 2 π 0 π¬ 0 π 0meant to say smt queries not set sorry
15.06.2025 19:05 β π 2 π 1 π¬ 1 π 0pulse: fstar-lang.org/tutorial/boo...
they also do some stuff with eliminating quantifiers (but I think that may be standard?)
okay so ive been thinking about this a bit more and one thing that is interesting is that pulse allows for introspection when writing proofs by having very targeted set queries, rather than discharging proof obligations all at once. idk if this is something that you looked at but maybe interesting ?
15.06.2025 18:54 β π 1 π 0 π¬ 3 π 0I see I see, makes a lot of sense!!
09.06.2025 00:22 β π 1 π 0 π¬ 0 π 0something I thought was interesting was that the guards that you mention already exist in verus :0 they have a pretty neat trigger system that they expose in the language which prevents a lot of the stability issues mentioned here. might be something to look at?
08.06.2025 23:25 β π 0 π 0 π¬ 1 π 0but why even accept that you HAVE to get mediocre output? this is one thing i never understood with vibe coding, like why not get it right the first time? why not actually put in the time and effort to make it good. if itβs something you want to make shouldnβt it be something youβre proud of?
04.06.2025 07:26 β π 3 π 0 π¬ 1 π 0A guide to programming paradigms:
Imperative: functions can have side effects
Functional: functions must be pure
Reversible: functions must be pure and bijective
Analytic: functions must be pure, bijective and infinitely differentiable
Agile: functions must take part in daily standups