Archive
Most engineers are trying to fix AI with better prompts. The Claude docs point to a different approach
Spec-driven development is quietly becoming one of the most powerful workflows in modern engineering
Most simple AI agents fail for one reason
Three years ago, I had a simple thought. Today, it is a reality
AI is generating more code than ever
Security Considerations When Integrating LLMs
The software engineering lifecycle has shifted, and code is no longer the hardest part.
Most real-world AI agent problems are coordination problems, not intelligence problems
Coding agents did not make product building faster. They moved where the time goes
Everyone is reaching for vector search. Few stop to ask if it is actually the right tool
Most RAG systems fail for a reason no one talks about
Most engineers use “vector index” and “vector database” as if they mean the same thing
I read a great piece on context engineering for AI agents, and it changed how I think about agent design
Unlearning becomes essential as you grow into senior roles
What I learned after 3 years of working with LLMs