# Hi, I'm Lenz. In one *hand-wavy* line, I work on debugging the coordination problems that make AI governance vulnerable outside the frontier. Currently, I'm the Co-convenor and Lead AI Risk Researcher at [AI4PH](https://ai4ph.substack.com/about), where I work on AI governance and workforce transition policy reforms in the Philippines. I'm also a regular facilitator at [Bluedot Impact](https://bluedot.org/) where I help run the [Technical AI Safety course](https://bluedot.org/courses/technical-ai-safety). Every now and then, I contribute to some research projects that involve building datasets for AI governance. You can [[Now|read more about what I'm working on here.]] I'm also interested in middle-power governance in Asia and in building the field infrastructure for AI safety work (e.g., replicating a bunch of papers, maintaining open-source repos for AI safety, building databases and datasets to support research, and taxonomizing). Previously, I worked on related projects at: - [ARENA 6.0](https://arena.education/), where I [debugged Inspect's CTF evals](https://docs.google.com/document/d/1BDodl_Y-KmnIQr4bjYVW1ra3IUw5eCFG4vf6JnOewh4/edit?tab=t.0#heading=h.r53rdq4qsot7), - [AI Safety Camp]([https://www.aisafety.camp/](https://www.lesswrong.com/posts/3sjtEXzbwDpyALR4H/ai-safety-camp-10-outputs)), where I theorized a [benchmark for value alignment](https://youtu.be/HabbyHTyKKk), - [Project GRASP](https://aisafetypreparedness.com/), where I curated literature on AI control, - [MIT AI Risk Repository](https://airisk.mit.edu/), where I helped [map the AI governance landscape of the U.S.](https://airisk.mit.edu/blog/mapping-the-ai-governance-landscape-pilot-test-and-update), and - [Condor Camp](https://global.condorinitiative.org/), where I ran Southeast Asia's 1st AI safety camp. --- [CV](https://bit.ly/ld_cv) · [Substack](https://halfbakedtheories.substack.com/) · [LinkedIn](https://www.linkedin.com/in/llenzl/) · [GitHub](https://github.com/ramennaut) · [Feedback](https://www.admonymous.co/lenz)