Lei is an Assistant Professor at the Technical University of Denmark. Lei received a Ph.D. in Computer Science, specializing in Mathematical Optimization, from Uppsala University in 2019. During the Ph.D., Lei was a visiting data scientist at BCG Gamma. Afterward, Lei worked as a data scientist at Bolt and Wolt in on-demand logistics optimization.

Lei develops verifiable AI systems: methods and information interfaces that make AI claims, decisions, and system changes independently checkable. Lei studies what evidence must be exposed, preserved, or tested as models, tools, memories, permissions, or environments change. This work examines verification in heterogeneous AI ecosystems, learning when human or automated checks are scarce, and explanations that account for data geometry, distributional structure, and fairness constraints. It draws on causal inference, information theory, and optimization to characterize what can be verified under limited evidence and verification resources. Lei is a co-founder of CSPaper, an open research infrastructure and testbed for studying verification-limited AI systems.