Josselin Perret@josselin

PolyM.AI/rket turns any binary question into a structured intelligence briefing : the kind that used to require a team of human analysts. Type a question. The system researches it live, then assembles a panel of 6–8 expert agents generated specifically for that question: a Fed chair for monetary policy, a RAND strategist for geopolitics, a VC and a regulator for tech governance. No fixed pool : the right experts for the right question, every time. The panel debates in two rounds. Round 1: each agent reasons independently from a shared fact sheet, outputs a probability and argument. Round 2: agents see only the consensus price — not each other's positions — forcing genuine conviction over social conformity. Aggregation uses logarithmic pooling (not a simple average), which corrects for overconfidence and produces a calibrated 80% confidence interval. On resolved historical questions, the ensemble scores a Brier score of 0.091 vs 0.148 for the best single agent — a 38% improvement. The final synthesis names who disagreed, why, and what new information would move the price. Built on AMD MI300X via AMD Developer Cloud. Qwen3-27B-FP8 on ROCm + vLLM serves 6–8 concurrent agents in parallel without memory pressure. Orchestrated with LangGraph, grounded with Tavily live search, streamed to a custom Slack-clone interface in real time.

Grab your ideas, spin up compute, and start building what’s next! Start with $100 in AMD Developer Cloud credits and begin building right away. Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further. To unlock credits and join the hackathon, sign up for the AMD AI Developer Program. 🤝 Go solo or team up — ship something real. 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs. 📍 On-site Venue (May 9-10): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 📍On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑‍💻 Developers, builders, and innovators - bring your boldest ideas and let’s turn them into real AI systems on AMD.