sreedev
banner
sree.dev
sreedev
@sree.dev
Software Engineer.
Radio Amateur // KD2ZWB
Brooklyn, NY.
www.sree.dev | www.devtechnica.com

0x{B8C4 02B1 6E80 E17C}
A potential solution to this would be to introduce some form of proof of stake requirement. The submitters must be required to pay a certain sum to make a submission that will be returned to them along with the promised bounty. If the report is found to be fake, the original sum can be forfeited.
August 5, 2025 at 10:25 PM
went right over my head haha. my bad.
December 11, 2024 at 1:21 AM
idk if you meant it sarcastically but there are some cool projects like supercollider and sonic-pi.

i recently also dicovered faust.grame.fr
December 11, 2024 at 12:51 AM
the only O(1) solution. haha 😂
December 9, 2024 at 4:48 PM
@tjdemuyt.bsky.social Thank you for that suggestion. Yes I did Fix that issue by using a custom type for each block that has an ID field. But the problem is the time complexity. I gotta try a different approach altogether.
December 9, 2024 at 4:11 PM
@ked1108.bsky.social that's awesome. go for it. Implementing libraries in rust is fun. I implemented an SDK for pocketbase a while back github.com/sreedevk/poc...
GitHub - sreedevk/pocketbase-sdk-rust: Rust SDK for Pocketbase
Rust SDK for Pocketbase. Contribute to sreedevk/pocketbase-sdk-rust development by creating an account on GitHub.
github.com
December 9, 2024 at 3:04 PM
A 100%. I thought the antenna antinode problem was going to be the ultimate edge case city (that turned out fine) until I saw today's problem
December 9, 2024 at 2:05 PM
@ked1108.bsky.social yeah I agree with you, I did reach for rust multiple times during AOC this year, but I wanted to just stick with gleam to learn it. I implemented (badly of course) a lot of the functions that other languages typically of provide in the stdlib
December 9, 2024 at 2:02 PM
@ked1108.bsky.social it brings me peace knowing I am not the only one haha. I am trying other approaches now.
December 9, 2024 at 1:47 PM
By only placing obstacles on the original path (Part 1), I was able to reduce the execution time to 1m. still working on optimizing it further.
December 6, 2024 at 6:15 PM
My Solution for Day 6 Part 2 took 9m and 30s to run, but it worked! parallelization worked!

Thank you 14 cores and 20 threads.
Run the solution at your own risk, it may or may not leak memory.

github.com/sreedevk/adv...

#AdventOfCode #Gleam
December 6, 2024 at 4:23 PM
Day 6 part 2 brute force works on the example but did not finish running on the input even after 2 minutes.

github.com/sreedevk/adv...

I could optimize the solution like a sane person, or I could use gleam/otp tasks and parallelize my problems.

parallelization it is!

#AdventOfCode #Gleam
December 6, 2024 at 2:27 PM
I did have to add a few list utility functions to make the solution look better. I didn't see any equivalents in the stdlib

github.com/sreedevk/adv...
advent-of-code/src/utils/list.gleam at main · sreedevk/advent-of-code
Advent Of Code Solutions. Contribute to sreedevk/advent-of-code development by creating an account on GitHub.
github.com
December 5, 2024 at 2:01 PM
saved by a “.pub”
December 3, 2024 at 4:23 PM
i too rolled out my own list.combinations lol

bsky.app/profile/sree...
December 2, 2024 at 8:49 PM
Looking at some of the other solutions, i just realised that I can replace the entire variants function with just

list.combinations(report, list.length(report) - 1)
December 2, 2024 at 2:33 PM