I enjoyed hearing your story! However, there were a few points that I found a bit disappointing. 1. I was expecting to see specific, practical examples of how you utilize agentic coding at a professional level, but the discussion felt a bit too high-level. As a result, it was hard to visualize what it actually looks like to work in an "AI Native" way. 2. I wanted to hear about your experiences regarding how to increase the reliability of AI-generated code. As a frontend engineer, I actively use agentic coding in my work, but I don't use it so aggressively that I can't review it myself. Since I occasionally see it implement things in overly complex or incorrect ways, I don't fully trust AI-generated code 100% yet. I think it would have been better if you had shown a bit more of how you solved this issue.