1.
2.
Nvidia says it can shrink LLM memory 20x without changing model weights
(venturebeat.com)
5.
6.
Prompt-caching – auto-injects Anthropic cache breakpoints (90% token savings)
(news.ycombinator.com)
7.
Ceno, browse the web without internet access
(news.ycombinator.com)
8.
What we learned from a 22-Day storage bug (and how we fixed it)
(news.ycombinator.com)
9.
Understanding the Go Runtime: The Scheduler
(news.ycombinator.com)
10.
Bcachefs creator insists his custom LLM is female and 'fully conscious'
(news.ycombinator.com)
11.
14.
Running AI models is turning into a memory game
(techcrunch.com)
17.
How DSQL makes sure sequences scale
(news.ycombinator.com)
18.
Expensively Quadratic: The LLM Agent Cost Curve
(news.ycombinator.com)
19.
20.
21.
Show HN: HN Companion – web app that enhances the experience of reading HN
(news.ycombinator.com)
22.
Nintendo 3DS emulation is getting a big performance boost
(androidauthority.com)
23.
Three Cache Layers Between Select and Disk
(news.ycombinator.com)
25.
26.
How to Scale a System from 0 to 10M+ Users
(news.ycombinator.com)
27.
28.
30.