I had F5Bot set up for keywords, but it was sending me 50+ alerts a day — mostly irrelevant. I'd spend 20 minutes triaging to find 2 threads worth responding to. The problem wasn't lack of monitoring, it was lack of signal.
Bruce is different: instead of matching keywords, it scores every item 0-100 against your product context, ICP, and competitor list. It understands why something is relevant, not just that a keyword appeared.
The part I'm most excited about: it learns. Every time you act on a signal or dismiss one, Bruce adjusts. By week 2, it's surfacing very different (better) things than week 1.
What it monitors: Reddit, Hacker News, RSS feeds, ProductHunt, and anything with an RSS endpoint via RSSHub (YouTube channels, newsletters, etc.)
Stack: Next.js 15, PostgreSQL, Drizzle, Better Auth, my self-built AI runtime, RSSHub.
Live at https://smartbruce.com
Happy to answer questions about the scoring model or architecture.
Show HN: Bruce – AI signal radar for Reddit/HN that learns what matters to you
smartbruce.com