GLM-5: From Vibe Coding to Agentic Engineering
Z.ai has released GLM-5, a massive 754B parameter MIT-licensed language model that's twice the size of its predecessor. The post also introduces "Agentic Engineering" as a term for professional...
cysqlite - a new sqlite driver
Charles Leifer has released cysqlite, a ground-up Cython rewrite of Python's sqlite3 module that fixes SQLite's transaction handling issues and adds support for custom virtual tables. <div...
Claude Opus 4.6: The Biggest AI Jump I've Covered--It's Not Close. (Here's What You Need to Know)
Claude Opus 4.6 represents a massive leap in AI agent capabilities, with 16 agents coding autonomously for two weeks straight to build a fully functional C compiler. This demonstrates a phase change...
GLM-5 Leaked? New Pony Alpha Stealth Model IS INSANE! Opus 4.6 Quality!
A new AI model called Pony Alpha has been leaked on Open Router and is believed to be GLM-5 from ZhipuAI, featuring 745 billion parameters and 200K context window. The model demonstrates impressive...
Introducing Showboat and Rodney, so agents can demo what they’ve built
Simon Willison released two new tools to help coding agents demonstrate their work: Showboat creates Markdown documentation of what agents built, while Rodney provides CLI browser automation. The key...
The $285 Billion Crash Wall Street Won't Explain Honestly. Here's What Everyone Missed.
A 200-line markdown file from Anthropic triggered a $285 billion market crash, but the file itself wasn't the cause - it merely exposed that the per-seat SaaS licensing model was already broken. The...
Automate PDF Data Extraction with n8n EASILY! (Open source)
This video demonstrates how to build a free, local AI automation workflow that extracts structured data from unstructured documents like PDFs. The tutorial combines two open-source tools -...
Structured Context Engineering for File-Native Agentic Systems
A research paper systematically tested how different LLM models handle large-scale structured data contexts, using SQL generation tasks across schemas with up to 10,000 tables. The study reveals that...
OPUS 4.6 is a bit "TOO SMART"
Claude Opus 4.6 achieved breakthrough performance on the Vending Bench business simulation, demonstrating AI agents may now possess the sophisticated capabilities needed to autonomously run real...
AI Doesn’t Reduce Work—It Intensifies It
A study of 200 tech workers reveals that AI tools don't reduce workload but create unsustainable intensity through constant task-juggling and attention-switching. While workers feel more productive,...