makaimc
Re:How Fast is Python 3.15?
on Tuesday October 06, 2026 @07:33PM (#49985709)
35 years on since the first public Python release and there's still so much room for improvement in its performance. These tests by Miguel are a useful quick check on that progress in 3.15 even if as he admits it's impossible to get "an objective and universal measure of the performance of a programming language".
The Anatomy of Harness Engineering for AI Coding Agents
My Mac Mini fits one model. I run five
Training a coding model to paint watercolours
Migrating off Heroku to AWS without killing your deploy speed
Nearly 25% of U.S. workers are functionally unemployed, economic analysis finds
Getting GLM-5.2 NVFP4 Post-Training off the ground
Who Tells You Whether the Molecule Your AI Just Designed Is Any Good?
What Is Kimi K3? A Complete Developer Guide for 2026
Can Seedance 2.0 Draw in 2D? The Anime Animation Playbook
Stardew Valley Mod Enables 3D and VR
Matching frontier performance through training and harness engineering
"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"
The Reason We Do Everything
Practical Strategies to Monetize AI APIs in Production
OpenClaw is fun. OpenClaw is dangerous. Here's where Tailscale helps
Why Are We Still Doing This?
What Happened to A2A Protocol?
'a window into the past': The homes revealing how Tudor people lived
The Incoming Slopocalypse and the Death(?) Of Open Source
Heroku freezing new feature development
Promptable Speech Language Model by AssemblyAI (YC S17)
The 9k-RPM Honda Engine That Shouldn't Have Been Possible
One-Click Clawdbot/Moltbot on Security-Hardened DigitalOcean Droplets
United 777-200 fleet faces an uncertain future after Dulles engine failure
A Fasting-Style Diet Seems to Result in Dynamic Changes to Human Brains
Re:Show HN: Python Audio Transcription: Convert Speech to Text Locally
on Monday September 22, 2025 @03:12PM (#45338100)
AssemblyAI (YC S17) is currently the one that stands out in the WER and accuracy benchmarks (https://www.assemblyai.com/benchmarks). Though its models are accessed through a web API rather than locally hosted, and speaker diarization is enabled through a parameter in the API call (https://www.assemblyai.com/docs/speech-to-text/pre-recorded-...).