Advent of Code in Go

It’s beginning to look a lot like Christmas!

⚠ To execute my solution(s) of any puzzle on Linux, replace all instances of \r\n by \n in all usages of the strings.Split function.

Execute 2025 solutions

After adding your puzzle input as input.txt in the directory of the solution, press CTRL+SHIFT+B to execute the VSCode Build Task.

Alternatively, you can run the

go run .\2025\dayXX\

command in the root directory, replacing XX with the day you want to see the solution of (any day below ten is referenced as 0X).

Execute earlier solutions (2023 & 2024)

After adding your puzzle input as input.txt in the directory of the solution, run in a terminal located in root

go run 202[3-4]/dayXX/partX.go

replacing XX with the day you want to see the solution of (any day below ten is referenced as 0X).

Optimized solutions

Optimized versions of my solutions for the following puzzles can be found as ../partX_optimized.go files:

• 2024 day 5, part 2
• 2024 day 6, part 2
• 2024 day 7, part 1
• 2024 day 20, part 1
• 2025 day 1, part 2
• 2025 day 3, part 2
• 2025 day 7, both parts

These versions constitute a different enough approach to the initial solving method so that keeping both versions stays relevant. Changes can be found at the top of the optimized file.

Click on the tiles below to get to my code!

2024 - 48 ⭐ - Go

2023 - 11 ⭐ - Go

“Inaccurate” 2024 times (days when I started solving the puzzle after 6 am CET)

• day 1: Anim’Est made it so that I couldn’t easily do the puzzle earlier
• day 2: Post-Anim’Est sleep (+ I hadn’t made the decision to start seriously doing puzzles)
• day 3: Started at 6:25 (hadn’t fully commited yet)
• day 15: Started at 6:16 (alarm somehow didn’t go off — the panic was real)

“Inaccurate” 2025 times

Disclaimer

Any solution and code that you will find in this repositery has been written entirely by hand, usually on the very morning the puzzle was released. This means no AI chatbot, no Copilot inline suggestions… I firmly believe that using these kinds of tools is of great use in programming in general, but that they go against the spirit of algorithmic puzzle solving. This does not mean that I believe that any solution you have designed using LLMs is worth less than what people have done without the use of AI; I just think that they do not have their place in competitive puzzle solving like Advent of Code :3