Dtrue
Turn messy social debate into credible signal. A structured debate app that replaces chaotic threads with clear flow, fast outcomes, and trustworthy summaries.
I built Dtrue because I was tired of watching good arguments get buried in social media noise. Traditional platforms reward volume over quality, and there's never a clear outcome—just endless threads that go nowhere.
Dtrue fixes this by structuring debates from start to finish. Users join with a stance (agree/disagree), post focused opinions, and see real-time signals that guide toward closure. When a debate ends, you get an AI-generated summary, participation stats, and ranked top opinions—something you can actually act on and share.
What Makes It Different
- Structured opinions with AI scoring instead of chaotic threads
- Timers and closure mechanics that produce real outcomes
- Fair visibility—everyone can be heard, credibility builds over time
- Cryptographic hashing for opinion integrity and auditability
Built & Shipped
Full-stack mobile app (React Native + Expo) with NestJS backend, PostgreSQL, Redis, and BullMQ async pipelines. AI summarization using DeepSeek + HuggingFace embeddings, Qdrant vector search for topic discovery, deployed on AWS. Currently in closed testing, launching soon.