About

Himanshu Garg is a skilled Compiler Engineer at Qualcomm, with a strong interest in Computer Architecture, Machine Learning, and Digital Design. He holds a Master of Technology degree in Electrical, Electronics, and Communications Engineering from the Indian Institute of Technology, Bombay. Himanshu has 4.28 years of relevant experience in the industry. Currently, Himanshu is working as a Machine Learning Engineer at Qualcomm RnD Bangalore, where he is involved in the development of the first generation of Hardware Accelerators and Compilers. His responsibilities include working on the support of different deep learning frameworks like Tensorflow, ONNX, and Caffe2 on AIC100, optimizing kernel development on Hexagon Vector DSP for object detection algorithms, and developing various deep learning operators supported by existing frameworks. He is also involved in post-training quantization techniques for achieving floating-point-like accuracy. Before joining Qualcomm, Himanshu worked as an Undergraduate Student at the Indian Institute of Technology, Bombay, where he gained valuable experience in the field. He also worked as a Product Development Intern at Adobe, where he was responsible for developing a web page using Node.JS for labelling of collected images and pre-processing data to convert it into the required LMDB and HDF5 data format using a Python script. He also converted previously available SqueezeNet to regression by adding Euclidean loss, Concatenation, and silence layer and fine-tuned it with the ImageNet model using GPU-enabled PyCaffe. Himanshu's technical skills include ML, Deep Learning, Object detection, Object Detection, Javascript, Tensorflow, Web, NodeJS, and Python. With his strong background in computer architecture, machine learning, and digital design, Himanshu is a valuable asset to any team looking to develop cutting-edge technology.