Harry CJ Martin
115 posts
@HCJMartin
Impractically practical.
If you want to get real work done - actually ship and build a business with velocity - I don’t understand why you would waste time trying to build your own code factory from scratch. This is an about-face from my article where I explained how to build your own code factory. That’s out dated now and a total waste of time. It’s even dumber to try to work locally in parallel on multiple threads with port clash, worktree headaches, checkout drift and more. Ever since I started using @DevinAI, my shipping velocity (and dev happiness) has increased drastically. One of the killer features is the “Test App” feature which literally works out of the box. It screen records and narrates itself clicking through the feature. Then it identifies bugs surfaced during the tests, then proceeds to fix them. Then when the PR is ship-ready, I just type !land and it merges to main (automatically reviewing and fixing all findings). It’s seriously insane. And now with gpt-5.5 it’s even more ridiculous. All this plus dependable, hardened cloud VMs - I feel like you have to be a masochist to go back to trying to work on your Mac. If you’re coding on your local machine instead of in the cloud, you’re falling behind. Trust me. The only reason I dev locally now is if I have a new UI-heavy feature, and need super fast iteration and micromanagement of the agent. For that, the new version of Codex with built-in browser control + annotations/comments is really good. It’s like @Agentationcom from @benjitaylor but it’s all built in out of the box.
A2UI v0.9 is the new standard for portable Generative UI, allowing your agents to natively "speak" UI directly to your existing frontends. ✨ React, Flutter, and Angular support 🐍 Python SDK via pip install ⚡ High-performance streaming 🛠️ Design system integration Learn more via the blog: goo.gle/3Oug7bi
