Phillip Wang

Phillip Wang

Builder, Researcher, Thinker


Blog - Resume - Contact
Github Twitter Goodreads

Hi! I'm Phillip Wang, a recent CS grad from CMU who's figuring out to live his life. I am interested in virtual reality, urban planning, math, the environment, and artificial intelligence. I care a lot about my friends, and am actively looking for ways to spend more time with them, either through telepresence or working on things with them. I think a lot about technology and how it can positively impact our society. I love reading biographies because I think people are fascinating, and love learning about all the sorts of ways that people live their lives.

I am currently the CEO/Co-founder of Gather, where we are building virtual spaces for all sorts of gatherings, starting with conferences, offices, and universities. In the future, we plan to build a decentralized platform for the VR Metaverse. I spent a high-quality year after graduation working full-time with two of my closest friends on Siempre.

I like building things. I have previously worked on projects like Subtle Asian Matches (dating app with 20k+ users), SquareCraft (Producthunt), and RoboVote (featured on Wired). Otherwise, I've built a 3D holographic video chat, a walking stick for blind people at the range of 40 feet, a full-sized humanoid robot controlled through VR, and experimented with noise cancelling speakers. In almost none of these projects do I work alone!

(If you are a young technologist reading this and the above resonates with you, I would HIGHLY recommend checking out HackLodge!)

I spent most of my undergrad on AI research. I believe that AI will be one of the most transformatively impactful technologies in the next few decades, and worked on research that I believed would help realize real-world impact. I worked on the MineRL project, using Minecraft to research imitation learning and sample-efficient RL. We recently released MineRL, a large-scale imitation learning dataset of humans playing Minecraft, and got a paper accepted to IJCAI 2019 and are hosting a NeurIPS competition. I have previously conducted research in meta-learning with Brandon Amos and released the first open-source implementation of SNAIL, a popular meta-learning architecture. I spent a good year at Microsoft building NLP models for emails, and an even better summer at Cruise Automation doing Computer Vision.