Ahsan Raza

Focused on making AI agent evaluations trustworthy through rigorous verifier design, benchmark validation, and reinforcement-learning environments. Passionate about ensuring that evaluation rewards reflect real capability, helping build AI systems that are more capable, reliable, and honestly measured.