Posts
All the articles I've posted.
-
How System Prompts Differ from User Messages — Persona Consistency and Forcing JSON Output
A question I ran into while experimenting with personas in the Claude API: how is a system prompt actually different from just writing 'please do X' in a user message? I worked through this in terms of consistency, security, and output enforcement.
-
Claude API Doesn't Remember — Maintaining Context in Multi-Turn Conversations
The API is stateless. You need to accumulate assistant messages in the messages array for the chatbot to remember the previous conversation. I confirmed this through a direct comparison experiment.
-
Why Vercel Kept Showing Old Posts — The Empty Content Collection Trap
I traced down a phenomenon where the previous posts kept showing up in the Vercel deployment even after I deleted all sample posts and pushed.