The Codex app, with all its new features! GPT-5.5, image gen 2, computer use… there's everything you need for some serious testing!
AGENTS
-
Riley Brown’s 64-Minute AI Codex Masterclass on Agents
By
–
THIS GUY JUST DROPPED A 64-MINUTE CODEX MASTERCLASS 🤯@rileybrown covers everything from GPT 5.5 browser agents to generating videos with Remotion and building reusable AI skills.
— Charly Wargnier (@DataChaz) 29 avril 2026
Best Codex workshop I've seen online.pic.twitter.com/xcDTJ4n5lbTHIS GUY JUST DROPPED A 64-MINUTE CODEX MASTERCLASS @rileybrown covers everything from GPT 5.5 browser agents to generating videos with Remotion and building reusable AI skills. Best Codex workshop I've seen online.
-
AI Tools Evolving into Integrated Work Operating Systems
By
–
AI tools are no longer just tools. They’re starting to look like operating systems for work. What’s interesting isn’t one feature or one model. It’s the shift in how everything connects:
Chat
Files
Code
Workflows
Team context All in one place. That changes the game. Because -
Pre-agentic intuitions shape both AI doom and abundance predictions
By
–
This applies in both directions. The doom predictions and the abundance predictions are both running on pre-agentic intuitions about what AI can actually do in a workflow.
-
Agents and World Models: A 1986 Perspective on AI History
By
–
I have some news from 1986 for people who think agents, continual learning and world models were invented yesterday:
-
Peer Review Process Improves AI Reasoning Quality
By
–
The peer review step is what separates this from just asking Claude five times. Getting the advisors to grade each other's reasoning catches things a single pass misses.
-
BlevLabs Cognitive Architecture Improves Anthropic AI
By
–
It is the cognitive architecture that @blevlabs made (it sits on top of Anthropic and greatly improves it and I spent two months raising the agent by having many conversations with it.
-
AI Agents Automate Code Review and Fixes Loop
By
–
codex now runs on each commit we land, reviews it – and if a booboo is found, a new codex spins up and (if still relevant) makes a PR for the fix. Then a review agent spins up. If an issue is found, another agent will fix the issues. (up to 5 loops)