Saurabh Khanduja
Machine Learning Engineer at IKARA Vision Systems
Kaiserslautern, Rhineland-Palatinate, Germany
Summary
🤩
Rockstar🎓
Top SchoolSaurabh Khanduja is a Machine Learning Engineer with 11 years of experience specializing in computer vision, real-time multi-camera tracking, and production image/video pipelines. He has driven deployment-grade solutions—from a training data management tool for geospatial annotations at Terraloupe to perception safety metrics for autonomous vehicles at Edge Case Research—and currently leads multi-object tracking and emerging 3D reconstruction product work at IKARA Vision Systems deployed across 1000+ sites. His academic work at TUM on neural network interpretability and weakly supervised semantic segmentation informs a practical focus on explainability and robust model behavior. A pragmatic engineer who moves research into production, he also contributes bug fixes to well-known open-source tooling for media metadata extraction, improving GIF handling and extractor robustness.
11 years of coding experience
8 years of employment as a software developer
Master of Science - MS, Informatics, 1.4, Master of Science - MS, Informatics, 1.4 at Technical University of MunichBachelor of Technology (B.Tech.), Computer Science, First, Bachelor of Technology (B.Tech.), Computer Science, First at Indian School of Mines
Hindi, English
Stackoverflow
4,276reputation
880kreached
105answers
2questions
opencv
top-5%
imagemagick
top-5%
python
top-5%
Github Skills (19)
metadata10
java10
javas10
python9
opencv9
imagemagick9
openframeworks6
refactoring6
hough-transform6
intellij-idea6
gzip6
camera-calibration6
unix6
jpeg3
webp3
Programming languages (13)
Github contributions (5)
drewnoakes/metadata-extractor
Mar 2017 - Apr 2017
Extracts Exif, IPTC, XMP, ICC and other metadata from image, video and audio files
Role in this project:
Backend Developer
Backend DeveloperContributions:5 commits, 3 PRs, 14 comments in 1 month
Contributions summary:Saurabh focused on bug fixes within the GIF image format reader in the `metadata-extractor` repository. Their commits addressed issues related to the correct interpretation of global and local color tables, along with converting values to unsigned formats. The user also added error handling for unknown GIF block markers. These changes improve the accuracy and robustness of the metadata extraction process for GIF images.
quicktimemetadataepsvideojpeg
saurabheights/MultiTrackPro
Mar 2023 - Sep 2024
Software to play synchronized video from multiple camera and run object tracking across cameras - WIP
Contributions:1 review, 1 PR, 8 pushes in 1 year 5 months
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.