Nathan S. Liu

WORK HISTORY

EDUCATION

ABOUT NATHAN S. LIU

UC Berkeley Chemical Engineering and CS dual bachelor’s, working at the intersection of machine learning, physical systems, and real-world deployment. Interests include scientific ML, industrial optimization, and BCI/neuroengineering. Most passionate about building tools that combine technical depth with practical impact, especially in ventures translating advanced science into scalable products.Random thoughts:Engineering has always been about imposing order on the physical world. For most of history, that was enough — understand the materials, the forces, the constraints, and build something that works. The 21st century has not abandoned that foundation, but it has altered what \"works\" means. The systems engineers build today are not static objects but adaptive, learning entities that operate across scales their designers may never directly observe.What changed is computation — not as a tool, but as a partner in reasoning. Machine learning introduces a different mode of problem-solving: the system discovers relationships the engineer did not specify and makes predictions the engineer could not derive from first principles alone. This is a philosophical shift in what it means to design.Consider modeling a physical system. An engineer writing governing equations encodes human understanding into mathematical form. A neural network trained on data from that same system may capture dynamics the equations miss, but without explanation. Neither approach alone is sufficient. First principles without data produce elegant models that fail in the field. Data without principles produce black boxes that fail silently and dangerously. The engineer who holds both reasoning modes simultaneously is practicing something genuinely new.When a creation can learn autonomously, accountability becomes strained. Fulfilling that responsibility requires designing for transparency and treating deployment as the beginning of continuous ethical judgment, not the end. The 21st-century engineer is not simply a builder but an integrator — of disciplines, reasoning modes, and human context.To be an engineer in this century is to stand at a genuinely new frontier, where the tools we wield can discover things we cannot yet explain and build systems that evolve beyond our initial designs. That is thrilling and sobering in equal measure. The question is not whether we will use these tools — we already are. The question is whether we will use them with the wisdom, humility, and moral seriousness they demand. That, more than any technical skill, defines the 21st-century engineer.