Our Research

Research & Publications

Our research spans the core domains of AI and ML, especially Multi-Agent Systems, but is not limited to this. Publications and detailed findings will be shared here as our work progresses.

Technical Report2026

Scaling Reasoning Through Interaction: Multi-Agent Systems at Test Time

How coordinated communication between AI agents unlocks performance gains that independent sampling cannot reach. This technical report presents the mathematical framework, experimental results across five benchmarks, and engineering practices that enable hundreds of agents to reason together efficiently — achieving new state-of-the-art results while using compute more effectively.