Can I run AI locally?

Prompt-caching – auto-injects Anthropic cache breakpoints (90% token savings)

Shall I implement it? No

Launch HN: IonRouter (YC W26) – High-throughput, low-cost inference

Claude now creates interactive charts, diagrams and visualizations

Kotlin creator's new language: a formal way to talk to LLMs instead of English

Malus – Clean Room as a Service

Are LLMs not getting better?

I don't use LLMs for programming

Reliable Software in the LLM Era

Many SWE-bench-Passing PRs would not be merged

Don't post generated/AI-edited comments. HN is for conversation between humans.

Personal Computer by Perplexity

Elevated errors on login with Claude Code

BitNet: 100B Param 1-Bit model for local CPUs

Create value for others and don’t worry about the returns

Cloudflare Crawl Endpoint

Agents that run while I sleep

I built a programming language using Claude Code

Amazon is holding a mandatory meeting about AI breaking its systems

RFC 454545 – Human Em Dash Standard

After outages, Amazon to make senior engineers sign off on AI-assisted changes

PgAdmin 4 9.13 with AI Assistant Panel

Redox OS has adopted a Certificate of Origin policy and a strict no-LLM policy

Levels of Agentic Engineering

Two Years of Emacs Solo: 35 Modules, Zero External Packages, and a Full Refactor

No, it doesn't cost Anthropic $5k per Claude Code user

Microsoft Outlook app now showing paid spam/phishing ad's

Things I've Done with AI

Code Review for Claude Code

Is legal the same as legitimate: AI reimplementation and the erosion of copyleft

Agent Safehouse – macOS-native sandboxing for local agents

We should revisit literate programming in the agent era

SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via CI

Sem – Semantic version control. Entity-level diffs on top of Git

Will Claude Code ruin our team?

Why developers using AI are working longer hours

LLM Writing Tropes.md

Autoresearch: Agents researching on single-GPU nanochat training automatically

LLMs work best when the user defines their acceptance criteria first

this css proves me human

Anthropic, please make a new Slack

Claude Code wiped our production database with a Terraform command

OpenAI launches GPT-5.4: New model hits 83% on pro-level knowledge benchmark

A tool that removes censorship from open-weight LLMs

I dropped our production database and now pay 10% more for AWS

We might all be AI engineers now

The next generations of Bubble Tea, Lip Gloss, and Bubbles are available now

A standard protocol to handle and discard low-effort, AI-Generated pull requests

GLiNER2: Unified Schema-Based Information Extraction