Blog

Technical notes, tutorials, and thoughts.

Integrating an LLM in Next.js 16: streaming and Server Actions
4 min read

Integrating an LLM in Next.js 16: streaming and Server Actions

How to wire Claude into a Next.js 16 app without freezing the UI: Server Actions, token-by-token streaming, and the right server/client boundary.

  • #ai
  • #nextjs
  • #claude
Effective prompting for developers: from loose prompt to system
3 min read

Effective prompting for developers: from loose prompt to system

Stop writing prompts on intuition and start designing them like software: structure, context, examples, and evaluation.

  • #ai
  • #prompting
RAG for devs: what it is, when to use it, and when it's overengineering
3 min read

RAG for devs: what it is, when to use it, and when it's overengineering

Retrieval-Augmented Generation explained without the hype: how it works, when it solves a real problem, and when you're just making life harder.

  • #ai
  • #rag