A couple of people have asked me what some of the main themes were from @aiDotEngineer. I used copilot CLI to analyse the transcripts from t…

A couple of people have asked me what some of the main themes were from @aiDotEngineer. I used copilot CLI to analyse the transcripts from the livestreams to help me answer this. The main question being asked is How much human oversight should we have? At dinner I sat next to one engineer from a startup that is not writing or looking at any code and controlling all his coding agents from slack. On the other side was an engineer working at an enterprise where engineers are only just now turning on agent mode in VS Code and when they do they approve every action. I also found it super interesting that 60% of Vercels traffic is bots and that they see it as a feature. Some people talked about 'Dark Factories' and I learned from the awesome devs at Amp that they are shipping straight to production with no PRs or reviews. On the other hand I spoke to Christina and Armin from Earendil and their talk was about their experience of just shipping landing them in a tricky position with tech debt. Mario in his talk seemed more in the middle, and talked about where to just ship and which tasks probably need a human. Not tagging anyone because I want to write a long blog post and re-watch with notes most of the talks again myself. So many other themes were mentioned and I'll link the website with the full summary I deployed to Github pages below❤️ And that's a wrap! AI Engineer Europe 2026 has concluded. Our video crew did incredible work to capture the energy, enthusiasm, and positivity of this event -- but it still doesn't come close to being there. If you're engineering the future of AI -- we hope to see you at a future…

Source: Swyx (X) | 2026-04-13

Related