Hi everyone, Iâm Emilie đ
Iâm new here and slowly moving from âAI-curiousâ to actually building with it. My background is in web development and some software projects, which means I spend a lot of time thinking about how LLMs fit into real systems without breaking everything around them. I care less about flashy demos and more about things that are stable, debuggable and safe for real users. Right now Iâm experimenting with using AI as a supporting layer: helping with code, answering domain-specific questions, and acting as a structured assistant inside existing products rather than trying to make it âthe whole productâ. Iâm especially interested in logging, guardrails, and what to do when the model is wrong or confused.. basically, how to design around failure instead of pretending it doesnât happen. I joined this community to learn from people who are actually shipping things, not just talking about them. Iâm happy to share my experiments, mistakes and small wins along the way, and Iâll probably ask a lot of questions about architecture, safety and practical use cases. Looking forward to learning with you all.