Computer science recap for week 2/2025

From 6/1/2025 to 12/1/2025

Articles #

• They squandered the holy grail 580 comments in 2 discussions

• Can LLMs write better code if you keep asking them to “write better code”? 478 comments in 2 discussions

• How We Cracked a 512-Bit DKIM Key for Less Than $8 in the Cloud 407 comments in 2 discussions

• How I program with LLMs 333 comments in 2 discussions

• Queues Don't Fix Overload (2014) 288 comments in 6 discussions

• Some Programming Language Ideas 261 comments in 2 discussions

• Flattening ASTs (and Other Compiler Data Structures) 211 comments in 9 discussions

• Zuckerberg approved Meta's use of 'pirated' books to train AI model 196 comments in 2 discussions

• Running Linux on an Intel 4004. The author wrote a MIPS emulator for a 4004 host! 188 comments in 4 discussions

• I quit my job to work on my programming language 188 comments in 5 discussions

• Bringing SerenityOS to real hardware, one driver at a time 156 comments in 2 discussions

• Pushing the whole company into the past on purpose 150 comments in 3 discussions

• Operating System in 1,000 Lines 130 comments in 3 discussions

• Building Ultra Long Range TOSLINK 123 comments in 2 discussions

• 8th 106 comments in 6 discussions

Releases #

• broccoli: A robust message queue system for Rust applications, designed as a Rust alternative to Celery 40 comments in 3 discussions

• Release UCM Desktop v1.0.0 2 comments in 1 discussion

Projects #

• LinearBoost: Faster than XGBoost and LightGBM, outperforming them on F1 Score on seven famous benchmark datasets 90 comments in 4 discussions

• Finally finished my Gameboy/Color emulator! 9 comments in 2 discussions

• byte: 6502 based fantasy console 8 comments in 2 discussions

• [P] Llama3 Inference Engine - CUDA C 5 comments in 1 discussion

• I'm compiling a list of JavaScript implementations of the Mandelbrot set. Contributions are welcome 2 comments in 1 discussion

• [R][P] distillKitPlus: High Performent Knowledge Distillation for LLMs 2 comments in 1 discussion

• [P] I built a library that builds tensors from reusable blueprints using pydantic

• Big-Endian RISC-V

Videos #

• Why L1 Regularization Produces Sparse Weights 7 comments in 2 discussions

• Stemming | Natural Language Processing | Easy To Understand 2 comments in 1 discussion

• [R] Seminar on Unlocking State-Tracking in Linear RNNs Through Negative Eigenvalues