Building AI systems to make computers, robots, and drones see and understand the world better.
Experience in embodied/spatial systems, distributed LLM training, human-LLM interactions, and diffusion image generation.
Has working Chinese fluency and provides a fast learning curve.
Selected Works
How to Convert Audio to Spectrogram Images | Make audio data visible to vision models.
Implementing and Optimizing Meanshift Clustering | Speeding up meanshift clustering by 6x with a GPU.
Decomposing Problems and Working with Data will get us Working Robots | My thoughts on how we can actually get working robots.
(Un)successfully Implementing DiffEdit | A paper from scratch implementation to edit images with text.
LLM Agents are While Loops | Implementing an agent with a simple Python
whileloop.Building SolveIt in SolveIt | Rebuilding SolveIt, an agentic LLM inteface that keeps the human in control, in SolveIt
Implementing a Neural Network from scratch | Progressively reimplementing PyTorch
Backpropagation Explained using English Words* | Explaining backpropagation simply enough so that a middle school student can understand it.
Implementing Stable Diffusion from its Components | Implementing Stable Diffusion by piecing together various models that make image generation possible.
