Alex Tianyi Xu

About Me

I am a Machine Learning Engineer at Tesla AI, where I work on Optimus. Before that, I worked on training and evaluation for FSD and Robotaxi.

I completed my Master’s in Computer Science at Carnegie Mellon University, where I was co-advised by Professor Reid Simmons and Professor Chenyan Xiong.

My research there spanned LLM domain alignment for low-resource programming languages, applied to reconstructing safety-critical traffic scenarios, and retrieval-augmented generation (RAG).

I previously also worked on low-resource knowledge distillation with Professor Louis-Philippe Morency’s MultiComp Lab, and social norms modeling with the Reliable Autonomous Systems Lab (RASL), tracking norm violations in real time with reinforcement learning agents.

Research Interests

• Embodied AI and Robotics: learning for humanoid manipulation, autonomy in the real world
• World Modeling: video generation, physics-informed world models
• LLM Alignment and Adaptation: domain alignment, low-resource adaptation, few-shot finetuning
• Machine Learning: deep learning, foundation models, knowledge distillation, reinforcement learning

Manuscripts

Outside of Research

Away from the keyboard, I am usually out with a camera. Check out my photos! →

Powered by Jekyll and Minimal Light theme.