juli1

Apple’s recent innovation bets — from a car to a VR headset to an AI-native product — have all flopped. Where is Apple headed? juli1.substack.com/p/where-is-app…

How I use AI today in my software development workflow juli1.substack.com/p/the-ai-devel…

Much of our critical infrastructure relies on old, outdated systems. I admire engineers who built stuff that lasts. juli1.substack.com/p/build-stuff-…

In a world where model performance changes week over week, you have to build eval very early and constantly run evaluation to see if a change is improving your tool accuracy or performance. Improve prompt -> run evals -> accept or reject changes Rinse and repeat

I have been looking more deeply at coding agents and spent some time looking at Aider. The biggest takeaway I have today is that all tools are doing only one thing: improving the prompt and the underlying tools fed to the LLM. juli1.substack.com/p/anatomy-of-a…

Understanding how coding agents are built and starting to dig into opencode implementation. juli1.substack.com/p/anatomy-of-a…

Codiga joins Datadog. Time to open a new chapter. codiga.io/blog/codiga-jo…