A rule I've come to trust in AI-assisted development: the model that wrote the code shouldn't be the one approving it.
So I built a Claude Code plugin that has OpenAI Codex review every commit Claude makes. P1/P2 findings block the push and send Claude back to fix its own work; clean reviews go through. There's also on-demand review for branches, uncommitted changes, and even implementation plans before any code is written.
Been running it privately for months, refactored and open-sourced it this week. Plain bash + jq, 49 tests, CI on Linux/macOS, MIT.
https://t.co/uUBgWewO4z
TensorFlow.js: train and use deep learning models directly in the browser, in JavaScript. Includes the full Keras API, and ability to load saved Keras models (and even fine-tune them in the browser)! https://t.co/SxUQ7xWf1A
This is to AI as prestidigitation is to real magic.
Perhaps we should call this "Cargo Cult AI" or "Potemkin AI" or "Wizard-of-Oz AI".
In other words, it's complete bullsh*t (pardon my French).
Tech Insider: you are complicit in this scam. https://t.co/zhUE4V2PSR
What are your favorite books related to Artificial Intelligence? 📚 Check out this thread in our forum for some interesting reads or join in the discussion! We’re always excited to add new books to our list. https://t.co/2dgChKmJYi