Summary
Steven Wang is a machine learning engineer and researcher blending hands-on ML tooling with rigorous research, currently a Machine Learning Intern at DeepJudge. He holds a BS in Electrical Engineering and Computer Science from UC Berkeley and an MS in Computer Science from ETH Zürich, and is based in Zurich, Switzerland. He previously contributed as a ML Researcher at The Atticus Project and as ML Research Engineer at the Center for Human-Compatible AI, where he led the development of the PyTorch/TensorFlow imitation learning library used in multiple publications and built 4 OpenAI Gym environments for the MineRL BASALT competition, including managing the MineRL dataset pipeline. He is an active open-source contributor with work across C++, Go, and Python—ranging from Layer 7 load balancing and network tooling to contributing to tcpdump, gopacket, BESS, NetBricks, and Kelda. Notable outputs include releasing and benchmarking MAUD (over 50k expert annotations, under review at EMNLP) and a probabilistically safe robot planning publication from UC Berkeley's InterACT Lab, underscoring his ability to translate research into reusable, production-ready tooling.
9 years of coding experience
7 years of employment as a software developer
Bachelor of Science (BS), Electrical Engineering and Computer Science, Bachelor of Science (BS), Electrical Engineering and Computer Science at University of California, Berkeley
Williams-Mystic Maritime Studies Program
Masters in Computer Science, Masters in Computer Science at ETH Zürich