👋 Hello there!
About Me:
👦 Name:
🅰️ First Name: Hong Kiat (HK),
🅱️ Last Name: Tan,
🆎 Listed Name: Hong Kiat Tan.
🍵 “Drinks/US” Name: Max.
👨👨👦👦 Pronouns: He/Him/His.
👨🎓 I am graduating with a Ph.D. in Mathematics from UCLA, advised by Andrea Bertozzi.
📚 I did my undergraduate studies at the National University of Singapore majoring/minoring in:
📗 Primary Major: Applied Mathematics,
📕 Secondary Major: Physics,
📘 Minor: Statistics.
Research Experiences & Interests:
My current research interests are in AI safety, and these include:
🤝 Multi-Agent Safety & Interpretability:
- Detecting and understanding collusion between AI agents (e.g. Schelling point, tacit, and acausal collusion).
- Applying techniques from game theory and decision theory.
- Adapting single-agent black-box and white-box monitoring methods to multi-agent systems.
🔬 Applied White-Box Methods & Model Organisms:
- Monitoring models through their internals as a stand-in for chain-of-thought (CoT) monitoring as CoT becomes less faithful.
- Building model organisms to evaluate the effectiveness of black-box and white-box methods.
- Developing and applying nonlinear/geometric methods for interpretability (drawing on neural geometry, Riemannian geometry, and differential topology).
🔭 Scalable Oversight via Interpretability:
- Developing foundational interpretability models/agents that can audit other models at scale, as a white-box counterpart to black-box protocols such as debate.
These have evolved from my earlier work in:
- 📊 Statistics: Causality/Causal Inference and High-Dimensional Statistics.
- 📐 Mathematics: Hyperbolic PDEs and Conservation Laws.
- ⚛️ Physics: Nuclear Astrophysics, Mathematical General Relativity.
Throughout, I think of myself as a “full stack mathematician” — I take projects end-to-end, from theory and modeling through implementation and simulation to production-scale systems whenever possible!
📜 You can access my publications/papers here.
✉️ I am always happy to discuss the projects/publications I have worked on in more detail. If you have any questions with regard to any of my publications, feel free to contact me via my email below!
CV:
📃 You can access my resume/CV here (AI Safety) and here (General). (Updated Oct 2026.)
☕ I am currently located in SF/Bay Area, so feel free to reach out for a coffee chat!
I’m mentoring two SPAR projects with David Williams-King and Linh Le in Fall 2026! Project details:
- Model Organisms and Interpretability for Scheming and
- Modeling and Detecting Unspoken Collusion (Schelling Point, Tacit, and Acausal).
I participated in the Iliad Intensive (October 2026).
I worked as an Applied Scientist Intern at Amazon Search (Summer 2026).
I worked as a Research Fellow with SPAR on a mechanistic interpretability project (Spring 2026).
I worked as a Quantitative Research Intern at WorldQuant on their Intraday team (Fall 2025).
I worked as a Data Scientist Intern at Amazon on the Search Data Science and Economics team (Summer 2025).
💼 I am looking for a full-time research role, with a strong preference for AI safety. Feel free to contact me via my email below!
Teaching:
🧑💻 Teaching Experiences:
- You can access them here.
Contact Me:
📩 maxtanhk@math.ucla.edu
🏢 MS6161