Tesla Motors, Senior Machine Learning Engineer, Nov. 2016–Today
Texas Instruments, System Engineer, Machine Learning Lab, May 2014–Oct. 2016
- Deep learning library development: TI Machine Learning Library (C/C++/CUDA), Caffe, Tensorflow, Torch, MXNet, Theano
- Deep learning applications research and implementation: image classification, multiple object detection (R-CNN, Fast R-CNN, Faster R-CNN), scene understanding, stereo depth estimation, motion estimation
Georgia Institute of Technology, Center for Signal and Image Processing, Graduate Research Assistant, May 2010–May 2014
The Pop in your Job: The What drives you - What is your passion?
- Conducted in-depth statistical analysis of orthogonal frequency-division multiplexing (OFDM) time-domain signals in optical wireless communications
- Proposed various optimization methods, e.g., convex optimization, integer programming, iterative clipping, and delta-sigma modulation, to reduce the peak-to-average power ratio of OFDM time-domain signals, which significantly improved power efficiency
There are so many uncertainties in autonomous driving. While deep learning is such a powerful tool that is solving many problems and will be solving more problems, you would not expect any deep neural network can cover every single corner case. These corner cases, of course frustrating me, but meanwhile drive me every day.