Tushar Singhal
Staff Software Engineer at Qualcomm
San Diego, California, United States
Summary
🤩
Rockstar🎓
Top SchoolTushar Singhal is a Staff Software Engineer in San Diego with four years of focused experience building machine learning tools and mobile neural network demos at Qualcomm, backed by a hands-on background in camera hardware design and SoC optimization. He combines ML model compression work—contributing practical TensorFlow and PyTorch examples to the AIMET project—with system-level expertise in Android app architecture for on-device vision processing. His trajectory spans low-level RTL and PPA improvements that yielded concrete area and performance gains to higher-level ML tooling and architecture, reflecting a rare full-stack lens on embedded AI. Trained in Data Science & Engineering at UC San Diego and with a BTech in ECE, he blends academic rigor with production impact and a penchant for bridging hardware constraints and modern model optimization techniques.
4 years of coding experience
9 years of employment as a software developer
University of California, San DiegoBachelor of Technology (BTech) Electrical Electronics and Communications Engineering, Bachelor of Technology (BTech) Electrical Electronics and Communications Engineering at Indian Institute Of Information Technology Allahabad Higher Secondary PCM with Computer Science, Higher Secondary PCM with Computer Science at Gayatri Public School
English, Hindi
Github Skills (14)
net10
quantization10
compression10
pytorch10
quants10
machine-learning10
lossless-compression10
deep-learning10
tensorflow10
python10
compress10
model-compression10
deep-neural-networks9
neural-network9
Programming languages (2)
Github contributions (3)
quic/aimet
Jul 2021 - Aug 2022
AIMET is a library that provides advanced quantization and compression techniques for trained neural network models.
Role in this project:
ML Engineer
ML EngineerContributions:3 reviews, 6 commits, 12 PRs in 1 year
Contributions summary:Tushar primarily contributed to examples and implementations related to machine learning model compression and quantization techniques within the AIMET framework. Their work involved developing and improving examples using TensorFlow and PyTorch, showcasing the application of channel pruning and spatial SVD methods. The commits included updates to support TF2.x, fix accuracy issues, and address potential loss calculation errors. These contributions enhanced the usability and practical demonstration of AIMET's compression capabilities.
pytorchtechniquesdeep-learningpruningcompression
quic-tsinghal/aimet
Jul 2021 - Aug 2022
AIMET is a library that provides advanced quantization and compression techniques for trained neural network models.
Contributions:11 branches in 1 year
pytorchtechniquesdeep-learningcompressionmachine-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.