Claude
-
v1.4: The Hallucination That Almost Made It to Production
Recently, I have been working on some new things that I always wanted to catch up on. In this AI space, it always feels like you’re way behind on concepts! But behind the constant flow of inventions (and sometimes disguised reinventions), there are very important core concepts that are interesting to know in order to… Read more
-
v1.3: Three Days Arguing With an LLM About What Year It Is
I once built a data analysis agent. Simple stuff – query the database, answer questions about transactions and campaigns. User asks something, agent writes SQL, runs it, gives an answer. Worked great until I asked: “Show me our Q1 2025 performance.” The agent pulled the data perfectly. Then told me: “This is a forecast for… Read more
-
v1.1: Why does AI code „always“ fail in production?
AI accounts for over half the code produced in some organizations, yet 48% of AI-generated code snippets contain vulnerabilities compared to lower rates for human-written code. The industry narrative is clear: AI-generated code is fundamentally less reliable in production. But what if we’re blaming the wrong thing? AI-generated code doesn’t come with a self-destruct mechanism.… Read more
276AI 