Hacker News: flo_rhttps://news.ycombinator.com/user?id=flo_rHacker News RSShttps://hnrss.org/hnrss v2.1.1Mon, 15 Jun 2026 04:46:12 +0000<![CDATA[Show HN: Velyr – an AI agent that finds and fixes conversion leaks on your site]]>Article URL: https://velyr.io/

Comments URL: https://news.ycombinator.com/item?id=48525761

Points: 7

# Comments: 1

]]>Sun, 14 Jun 2026 10:00:55 +0000https://velyr.io/flo_rhttps://news.ycombinator.com/item?id=48525761https://news.ycombinator.com/item?id=48525761<![CDATA[New comment by flo_r in "Show HN: Paca – Lightweight Jira alternative for human-AI collaboration"]]>Gh issues works surprisingly well as an agent board. Labels for state, one issue per feature. The part i haven't figured out yet is how to know when the output is actually done vs just "looks done" to the agent.

]]>Sat, 13 Jun 2026 17:12:02 +0000https://news.ycombinator.com/item?id=48519244flo_rhttps://news.ycombinator.com/item?id=48519244https://news.ycombinator.com/item?id=48519244<![CDATA[New comment by flo_r in "Slightly reducing the sloppiness of AI generated front end"]]>Tried macOS HIG for the same reason and got similar results, less slop, more structure. I think what's happening is that the model has a very specific grammar to pull from instead of averaging over everything web-related it's ever seen. The SaaS one is interesting as a control, it's essentially asking for the average of all modern web UI, so you get exactly that.

]]>Fri, 12 Jun 2026 20:39:15 +0000https://news.ycombinator.com/item?id=48509179flo_rhttps://news.ycombinator.com/item?id=48509179https://news.ycombinator.com/item?id=48509179