Loading...

Tag trends are in beta. Feedback? Thoughts? Email me at [email protected]

A faster Easter date function

JIT bitmap text rendering for microcontrollers

Rust changed me as a developer

Shlemiel the Painter Is Back: How Zig’s libc Rewrite Made SQLite Slower

Optimizing Reference Deletions in Git

Writing a DSL for high performance parallel programming in OxCaml

Next-Level Benchmarking Overfitting – Thomas Neumann

Nix: Using the eval-profiler

11 Nanoseconds per Word: Over-Optimizing Anagrams

funkycache: a Python library to cache functions in SQLite

Scriptable GeoDNS: Leaving AWS for Bunny

The State of Python Profiling in 2026

A C++ Quicksort Faster Than std:sort

Non-Determinism and Overcount on Modern Hardware Performance Counter Implementations

Compiler performance, the next-gen trait solver, next-gen region constraints, implicit default auto traits, const generics, view types, and custom allocators: What we are working on at Hexcat…

Performance of a formally-verified C compiler vs. GCC -O2

hotpath-rs - Rust Benchmarking: How to Detect Performance Regressions in CI

After rewriting a small Python service in Rust, the real win wasn't performance

How to optimise compiler performance with very macro-heavy code?

Scheme+ language : 100% Scheme compatible, infix support and now with type annotation and inference for performance.

A study of over 80,000 online learners finds that internet delays of just 1–2 seconds are linked to lower academic performance, more distractions, and fewer completed lessons

APSR study: The election of mayors who are backed by police unions tends to lead to an expansion in police department size, but no improvements in police performance.

Performance evolution of route lookup on Linux

'Like gene editing for neural networks': MIT researchers used GPT-6 Astra to edit Qwen's open weights directly, boosting math performance

AiLang, Artificial Intelligence Programming Language with a Self hosting x86_64 compiler, performance comparable to gcc -03 llvm clang in many instances.

Show HN: NoDOM high performance WASM UI kit for DuckDB-WASM

Google expands its chipset lineup with up to 80% better inference price performance

Closed AI models dominate the price-performance frontier

Panthera: Ampere's CPU Performance Model

South Korea targets ' highest performance' sovereign AI model

More →