All activity
egedevleft a comment
We tested this setup on 100K+ real messages last week: • Avg. decision time: <480ms • Auto-approved: 82% • Flagged: 13% • Rejected: 5% What surprised us most — even simple rules like “Reject only if message targets a person” cut false positives by almost 40%. https://www.producthunt.com/products/moodiqo
I built a context-aware moderation rules engine — feedback on rule design?
egedevJoin the discussion
egedevstarted a discussion
I built a context-aware moderation rules engine — feedback on rule design?
I’ve been exploring a rules + LLM approach where you write plain-text rules (any language) and the system makes approve/flag/reject decisions in under ~500ms, with priority to resolve conflicts. What I’d love feedback on: • Is priority the right way to resolve rule clashes (e.g., Personal Attack 0.9 > Verified 1.0 auto-approve)? • Which starter rules would you expect for gaming, reviews,...
egedevleft a comment
Hey everyone 👋 I built Moodiqo because every moderation API I tried failed at understanding context. They detect bad words, but not the meaning behind them. Moodiqo lets you write simple natural-language rules like: { "prompt": "If a message praises politician A but criticizes politician B, approve it. If it praises B or criticizes A, reject it — even if the language is polite.", "action":...
MoodiqoBuild smarter filters, not word blocker