Hey hey, Useful AI Coding links to read/watch for this week of May 13, 2026.
If you want to get such list every Wednesday, subscribe via the form at the bottom of the website.
For Premium Members
6 Chinese LLMs: Coding Test on TWO Laravel Tasks
11-minute video. I have given two tasks to 6 different LLMs - Kimi, MiMo, Deepseek, GLM, Qwen, and Minimax. Is there a clear winner?
My YouTube Videos
6 Chinese LLMs: Coding Test on Laravel Task
Free version of ONE task of the premium video from above.
I Tested Grok 4.3 vs Other LLMs for Coding: Clear Answer
In the comments, you asked me to add Grok 4.3 to LLM comparison. I did, and let me show you the result.
I Tested 6 LLMs with MULTIPLE Runs for Same Prompt
If we launch the same prompt on the same LLM, would the result be the same? Of course, not. But how different? This is exactly what I wanted to test.
Planning with AI Agents: Markdown vs... HTML?
This tweet EXPLODED over the weekend in AI coding community. So let me explain and show more examples of what Thariq is ACTUALLY trying to say here.
AI Coding Community
Matt on X: "Two new skills this week!"
x.com
- /handoff compacts your current session to a markdown file - /prototype helps you prototype anything - UI or backend And a bunch more updates.
Claude on X: "New in Claude Code: agent view"
x.com
Run claude agents to start dispatching multiple sessions at once. Each one keeps running without taking up a terminal tab.
OpenCode on X: "DeepSeek V4 Flash - free for a limited time"
x.com
DeepSeek V4 Flash is currently our most popular model in Go Give it a try if you haven’t already
Codex just released remote control!!
x.com
If you’ve ever wanted to continue your codex session while touching grass, OpenAI is making it happen, and with the Codex iOS incoming. - start a codex session on your local machine - remote control - continue on codex iOS
- Kimi K2.6
- GLM-5.1
- ...
CJ on X: "Codex 5.5 hack: Are you 100% confident in this strategy?"
x.com
"Are you 100% confident in this strategy? If not, find all possible loopholes, suggest proper fixes and run this loop until you are factually 100% confident in the new startegy" This works like charm. It makes Codex 5.5 high perform even better than codex 5.5
This, along with our other recent compute deals, means that we’ve been able to increase our usage limits for Claude Code and the Claude API.
Daniel on X: "Claude Code 2.1.139 added /goal"
x.com
You set a completion condition and Claude keeps working across turns until it's met Works in interactive, -p, and Remote Control
Owl-Listener/designer-skills
github.com
Designer Skills Collection: agentic skills, commands, and plugins for design — from research to systems, UI, interaction, and delivery.
That's it for this week, see you in the next newsletter issue!
Again, if you wanna get this to your inbox every Wednesday, subscribe at the bottom!
