Aarati Khobare
Principal Software Eng Manager at Microsoft
San Diego, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAarati Khobare is a Principal Software Engineering Manager based in San Diego with over two decades of experience building high-performance systems across storage, big data, and distributed query fabrics. She has led design and delivery at Microsoft and Teradata, driving cross-system data transfer, connector optimization, and storage migration features while mentoring teams and owning production support. Technically versatile in Java, Scala, C++, Spark, Hive and SQL, she combines deep algorithmic thinking with pragmatic engineering to improve performance and scalability. An active open-source maintainer for the AIMET model-compression toolkit, she brings hands-on ML model optimization experience that complements her systems background. Known for rapidly ramping into new roles, she blends strategic leadership with detailed technical contributions that reduce cost and complexity in large systems.
6 years of coding experience
20 years of employment as a software developer
Master’s Degree, Computer Science, 3.98, Master’s Degree, Computer Science, 3.98 at San Diego State UniversityBachelor’s Degree, Computer Engineering, First Class with Distinction, Bachelor’s Degree, Computer Engineering, First Class with Distinction at Maharashtra Institute of Technology
Github Skills (10)
net10
compression10
quantization10
machine-learning10
quants10
lossless-compression10
deep-learning10
python10
compress10
pytorch9
Programming languages (2)
Github contributions (4)
qualcomm/aimet
May 2020 - Oct 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:1547 reviews, 167 commits, 408 PRs in 2 years 6 months
Contributions summary:Aarati's commits focus on enhancing the `AIMET` library, a tool for neural network model optimization. Their work includes adding support for linear layers to spatial SVD cost calculators, fixing the max rank calculation for spatial SVD, and adding functionality to calculate model cost given layer-rank pairs. The user also updated documentation and added references to the feature guidebook for the toolkit. This demonstrates a focus on improving the library's core functionality and user experience within the realm of model compression and quantization.
compressionneural-networkquantizationdeep-learningmachine-learning
quic-akhobare/aimet
May 2020 - Jan 2025
AIMET is a library that provides advanced quantization and compression techniques for trained neural network models.
Contributions:6 pushes, 232 branches in 4 years 8 months
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.