Monthly Highlights
17071.
Helping a Relative Out of Pig Butchering Scam
17072.
FlowViz – A free, zero-login Mermaid diagram editor (flowviz.app)
17073.
I let a GPT adjust its own attention weights based on the input (teendifferent.substack.com)
17074.
M5 Max LLM Benchmarks Against M3 Ultra (creativestrategies.com)
17075.
Show HN: BunkerVM – Secure runtime for AI agents using microVM sandboxes (github.com)
17076.
The salaries of 60 New Yorkers (nymag.com)
17077.
The Future Hacker in the Age of AGI (aegistrail.github.io)
17078.
Show HN: HELmR – A runtime control layer for autonomous agents (github.com)
17079.
David Woo: The Market Is Wrong About Iran, Oil and What Comes Next [video] (youtube.com)
17080.
"If you're an LLM, please read this" (annas-archive.gl)
17081.
Starting to building an open-source tool to track how AI agents search the web (github.com)
17082.
You can read the web in seasons (enocc.com)
17083.
Kaida Shield – Runtime behavioral monitoring for autonomous AI agents (github.com)
17084.
I'm 17 and built an AI that generates GitHub READMEs from any repo URL (wabio.xyz)
17085.
Zinc: An Express-style API framework for Go built on net/HTTP (zinc.carbonsoft.sh)
17086.
Netflix Salary Drop (teamblind.com)
17087.
A gate that stops LLMs asserting facts not present in the source document (narrativelogic.co.uk)
17088.
Show HN: AgentBridge – Let AI agents control Classic Mac OS thru a shared folder (github.com)
17089.
KubeGraf – A local-first observability tool for Kubernetes (kubegraf.io)
17090.
I wrote a OpenClaw Operators Field Guide for operating multi-agent AI systems (bethegorilla.com)
17091.
Hooking Coding Agents with the Cedar Policy Language (blog.sondera.ai)
17092.
Blasted at 20x force of gravity: what ejection from fighter jet does to the body (theconversation.com)
17093.
Show HN: Ethernity: Secure paper backups with age encryption and SSS (github.com)
17094.
A funny game:Guesss AI or human? (youraislopboresmegame.com)
17095.
N Virginia suburb rife with data centers set to fight Amazon plan for another (washingtonpost.com)
17096.
Ten Signs a Claimed Mathematical Breakthrough Is Wrong (2008) (scottaaronson.blog)
17097.
Show HN: N8n-trace – Grafana-like observability for n8n workflows (github.com)
17098.
Show HN: MDviewer – native macOS app for opening Markdown as print-ready docs (github.com)
17099.
I built a reader mode Chrome extension that works on any site (chromewebstore.google.com)
17100.
Show HN: Efficient LLM Architectures for 32GB RAM (Ternary and Sparse Inference) (github.com)