They made an AI become self-aware about spaghetti

Ramp's research team built a system that steers an LLM's personality at inference time by slapping a vector on top of every token prediction. The result? A model that starts talking about spaghetti mid-conversation — and then catches itself doing it. Alex Shevchenko explains how steering vectors work and why the model sometimes loops into existential crisis. This clip is from Max Agency, a #podcast about how the best AI agents are actually being built. Hosted by Harrison Chase, CEO of LangChain, each episode goes deep with the builders designing, deploying, and learning from real agent systems in the wild. From architecture decisions to evals, tooling, and failure modes, Max Agency is for people who want to understand what it really takes to build useful agents.