Daniel
Lee
|
Stanford University · Mathematics & Computer Science
2× AIME QualifierUSACO GoldMIT AI Hack-Nation Top 1.5%
Experience
Stanford AI Lab
ML Researcher · Jun 2025 – Present
- Developing interpretability methods to discover discrete reasoning operators in large language models.
Meta
Software Engineer Intern · Jun 2026incoming
Projects
OPUS
Discovering discrete reasoning operators in large language models via mechanistic interpretability.
PyTorchTransformersInterpretability
MoSV
Mixture-of-Steering-Vectors framework for targeted hallucination reduction in LLMs.
Steering VectorsLLMAlignment
DYNAMO
Reinforcement learning agent for dynamic portfolio optimization under market uncertainty.
RLFinancePyTorch
ShED-HD
Entropy-based hallucination detection framework leveraging token distribution signals.
EntropyUncertaintyNLP