What happens when a model only makes the decision and code does everything else? I experimented with Jev, a decision model from TypeSafe AI, and wrote up what it decides, what code keeps, and the practices that held up.
Building a chatbot with LangGraph that supports multiple languages is harder than it looks. Here is how we solved it without calling an LLM for every yes or no.
Practical examples of how I use Cursor (AI editor) and Warp (AI terminal) to refactor code, write tests, and speed up everyday tasks with prompts, examples, and images.