422
Claude Code sends 33k tokens before reading the prompt; OpenCode sends 7k
236
systima.ai
398
Old and new apps, via modern coding agents
113
terrytao.wordpress.com
289
I love LLMs, I hate hype
171
geohot.github.io
244
How to read more books
136
scotto.me
220
Since Chromium 148, Math.tanh is now fingerprintable to link underlying OS
109
scrapfly.dev
187
Irish datacenters now guzzle 23% of the country's electricity
151
theregister.com
179
Don't you mean extinct?
107
fabiensanglard.net
132
LARP – Revenue infrastructure for serious founders
31
larp.website
114
Tiny Emulators
3
floooh.github.io
103
Migrating a production AI agent to GPT-5.6: 2.2x faster, 27% cheaper
29
ploy.ai
93
Why write code in 2026
140
softwaredoug.com
82
Automation Without Understanding
37
arxiv.org
79
I Learned to Read Again
37
substack.magazinenongrata.com
77
Mechanistic interpretability researchers applying causality theory to LLMs
60
cacm.acm.org
76
Against Usefulness
21
motivenotes.ai
58
How we can reduce traffic congestion
47
research.google
58
Deir El-Medina Strikes
14
en.wikipedia.org
57
So you want to learn physics (second edition, 2021)
5
susanrigetti.com
42
Show HN: Agent Draw: An agent draws while you talk, built on TLDraw
15
techstackups.com
39
Ask HN: What Are You Working On? (July 2026)
87
news.ycombinator.com
38
The One-Step Trap (In AI Research)
7
incompleteideas.net
28
Kode Dot Programmable pocket device for makers, pentesters and geeks
6
kode.diy
24
Cyberpunk Comics, Manga and Graphic Novels
2
shellzine.net
15
Flash-MSA: Accelerating Million-Token Training with Sparse Attention Kernels
0
nanduruganesh.github.io
15
Show HN: Zen Mode – a global focus mode for macOS
6
github.com
13
Designing and assembling my first PCB
0
vilkeliskis.com
10
A Speed Limit for Computers
7
caolan.uk
10
Profiling the "Abundance" housing bottleneck with real data
3
laxmena.com
7
Architecture Description Languages [pdf]
1
ics.uci.edu
4
How to Get What You Want
0
andys.blog