M.Sc. Mechatronics student at FAU Erlangen–Nürnberg. Working at the intersection of deep learning, sequence modelling, and physics-constrained control — with current hands-on experience in sequence models (LSTM / temporal CNN / attention) and physics-informed losses.

My sEMG impedance work taught me that low-error AI predictions still get bolted to classical safety filters because no one can be accountable for a black-box decision. Causality is too brittle a remedy. I am now drawn to AI world models — multi-modal, physically grounded predictions whose future is observable and checkable from the prediction itself.

• Blog & longer thinking: tianhao1314.github.io
• Design-space study: semg-impedance-prediction
• Email: [email protected]

M.Sc. Mechatronics · FAU Erlangen–Nürnberg · Currently preparing thesis applications.