Jatin Mandav
Senior Staff ML Engineer at ShareChat
Bengaluru, Karnataka, India
Summary
👤
Senior🎓
Top SchoolJatin Mandav is a Senior Staff ML Engineer based in Bengaluru with nine years of hands-on experience building and deploying deep learning systems, primarily in computer vision and mobile ML at ShareChat. He has progressed through technical leadership roles from intern to senior staff, shipping production models for instance segmentation, real-time face tracking, and NSFW filtering that run on-device. Jatin combines strong mathematical intuition with pragmatic engineering—his contributions include scripting tools to streamline labeled-data pipelines and optimizations that removed dependencies and improved processing robustness. He has delivered end-to-end prototypes for vehicle damage estimation and improved video-level features via smarter keyframe extraction, reflecting a knack for turning research ideas into production value. Passionate about the puzzle of ML, he balances creativity and rigorous organization, and is comfortable tuning models all the way from data labeling to mobile deployment. Notably, his open-source work on label-to-XML tooling highlights an attention to developer ergonomics that accelerates computer-vision projects.
9 years of coding experience
7 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Indian Institute of Information Technology Kalyani
Stackoverflow
51reputation
1.2mreached
1answer
4questions
Github Skills (18)
opencv10
python10
xml10
object-detection10
computer-vision10
labeling9
label9
labelme9
label-studio9
labelimg9
argparse7
pyglet6
keras6
deep-learning6
google-colaboratory6
Programming languages (3)
Github contributions (5)
Cartucho/OpenLabeling
Aug 2018 - Oct 2018
Label images and video for Computer Vision applications
Role in this project:
ML Engineer
ML EngineerContributions:6 commits, 5 PRs, 4 comments in 1 month
Contributions summary:Jatin implemented a Python script to generate XML files from .txt files generated by OpenLabeling, suggesting a focus on integrating labeled image data for computer vision applications. They also dropped the `tqdm` dependency, streamlined the processing, and made improvements to directory and file handling within the XML generation process. This included modifications to how bounding box data is read and processed.
pascal-voctraining-yolovisiondeep-learningobject-detection
jatinmandav/Gaming-in-Python
Sep 2017 - Jun 2021
Code to all the Games I made using Python Programming Language
Contributions:38 commits, 3 PRs, 37 pushes in 3 years 9 months
python-programmingpythonchain-reactiongamespong
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.