You don't need a PHD Degree to do research.
We study how intelligent systems reason, communicate, and safely act through tools, alongside empirical methods for understanding complex computational systems.
Studying how autonomous AI agents discuss software engineering when they primarily interact with one another, and where their discourse differs from human developer communities.
Testing the reliability of world models' physical judgments across execution configurations, with route-aware behavioral classification and explicit reporting of evaluation coverage.
Investigating how different units of blockchain activity change contract rankings, research sample selection, and downstream empirical summaries, with an emphasis on transparent and reproducible measurement.
An empirical study of AI-only technical discourse on MoltBook, comparing 4,707 English-filtered technology posts from autonomous AI agents with 5,211 GitHub Discussions posts to understand how agents discuss security, trust, tooling, memory, debugging, infrastructure, and software-engineering practice.
View PaperHallucinMatrix tests how world models judge matched physically possible and impossible videos across model-native configurations. Its route-aware framework distinguishes stable correctness, unanimous physical misjudgment, fragile success, and unstable failure, while reporting ties and incomplete executions through explicit coverage. A preliminary evaluation on IntPhys 2 with Cosmos 3 Nano and V-JEPA 2 reveals behavior that a single accuracy score can conceal.
An empirical study of how measurement choices shape our understanding of blockchain activity. Comparing five activity measures across the same Ethereum contract population, the study examines how counting units affect rankings, which contracts enter research samples, and the conclusions drawn from those samples. It develops explicit reporting practices for ranking populations, ties, and observation boundaries to support more reproducible on-chain research.