Creating worlds out of thoughts.
Weekend experiment: can you generate virtual worlds directly from brain signals? Combined Nvidia Cosmos with DreamDiffusion to turn EEG readings into explorable environments.
I am currently a master's student at ETH Zurich.
I am interested in 3d vision, self-supervised learning and solving scalability for vision. I believe vision is the most important frontier with 70% of internet traffic being video and more than half of the human brain being devoted to vision processing.
Previously I completed my bachelors at ETH Zurich in electrical engineering, studied at NUS in Singapore and worked as a computer vision engineer at Polybee.
Weekend experiment: can you generate virtual worlds directly from brain signals? Combined Nvidia Cosmos with DreamDiffusion to turn EEG readings into explorable environments.
A tool to semantically search for researchers and find knowledge clusters based on co-authorship. Built in 24 hours for the Berkeley AI Hackathon.
A demo to explore how vector search could change marketplaces. Enables semantic product discovery, allowing users to search by concept rather than keywords.
Built at the dawn of the computer agent era in 24 hours and won Singapore's largest hackathon.
A novel architecture utilizing Monte Carlo Tree Search to generate interconnect topologies of diameter-3. This project, my bachelor's thesis, taught me first principles thinking and scalable engineering. The research on this project continues at the lab but this published paper uses my work.
This project taught me systematic ML experimentation: proper benchmarking, rigorous ablation testing, and drawing conclusive insights. Trained and evaluated a superresolution model based on this work on MRI images.
Once I started, there was no turning back. Spent 36 straight hours solving it.
I worked on the avionics stack of the 2023 BERNOULLI hybrid rocket. Launched at EUROC from a Portuguese army tank training ground.
My first real ML project. Trained a lightweight neural network, converted it to embedded C, and deployed it on an STM32 microcontroller for real-time song classification (~90% accuracy) using only a microphone and FFT-based features.
In my final year of high school, I became obsessed with understanding how China went from importing technology to dominating high-speed rail. Spent months diving into their strategy—tech transfer, absorption, leapfrogging. The thesis won provincial and national awards.