Arvind Ganesh Balasubramanian
GitHub LinkedIn Google Scholar

Arvind Ganesh Balasubramanian

Engineering Leader · Machine Learning

  1. Roblox 2026 – Present
    Engineering Manager Leading Ranking & Personalization for the Avatar Marketplace.
  2. Uber 2024 – 2025
    Tech Lead Manager Building enterprise search and LLM-powered productivity tools.
  3. Google 2014 – 2024
    Software Engineer & Tech Lead
    • Personal video search and the Action Moments GIF feature on Google Photos.
    • ML pipelines for ranking and moderating user-uploaded imagery on Google Maps.
    • Multimodal models rating ad creative quality on the Discover Feed.
    • LLM-based tools assisting developers with large-scale code migrations in Search.
  4. Baarzo 2013 – 2014
    Founding ML Engineer Video search startup acquired by Google in 2014.
  5. IBM Research 2012 – 2013
    Research Scientist Data clustering and imputation with partial information. Applied it to build tool to aid customer care reps with issue resolution as they gather new info from customers.
  6. University of Illinois, Urbana-Champaign 2006 – 2012
    Ph.D., Electrical & Computer Engineering Low-rank matrix recovery methods in computer vision. Advisor: Prof. Yi Ma
  7. Indian Institute of Technology, Madras 2001 – 2006
    B.Tech & M.Tech, Electrical Engineering

Arvind Ganesh Balasubramanian is an Engineering Manager at Roblox, where he has led the Ranking and Personalization team for the Avatar Marketplace since January 2026.

Arvind is an experienced engineering leader with deep expertise in machine learning, computer vision, and applied AI. He spent a decade at Google across a range of high-impact roles. He built ML pipelines for parsing and moderating user-uploaded imagery on Google Maps, developed multimodal models to assess ad creative quality on Discover Feed, and led the development of LLM-based tools to assist developers with large-scale code migrations. Earlier at Google, he worked on personal video search for Google Photos, deploying multiple models that significantly improved recall and grew query volume, and proposed the Action Moments feature for capturing interesting moments from videos as shareable GIFs. Before joining Google, he was a founding engineer at Baarzo, a video search startup that was acquired by Google in 2014. He also spent a brief stint as a Tech Lead Manager at Uber, building enterprise search and LLM-powered productivity tools.

Arvind holds a Ph.D. in Electrical and Computer Engineering from the University of Illinois, Urbana-Champaign, and a B.Tech & M.Tech in Electrical Engineering from IIT Madras. He is a Senior Member of the IEEE and a named inventor on two US patents in the areas of video understanding and synthesis.

Publications

Patents

2020
Generating a Video Segment of an Action from a Video

S. Vijayanarasimhan, A. Bienvenu, D. Ross, T. Novikoff, A. Balasubramanian

US Patent No. 10,740,620 · Filed 2017-10-12 · Granted 2020-08-11 · Assignee: Google LLC

2019
Generating Synthesis Videos

A. G. Balasubramanian, A. K. Kandoor, A. Raju

US Patent No. 10,410,060 · Filed 2017-12-14 · Granted 2019-09-10 · Assignee: Google LLC

Book Chapters

2012
Face Recognition by Sparse Representation

A. Ganesh, A. Wagner, Z. Zhou, A. Yang, Y. Ma, J. Wright

Compressed Sensing: Theory and Applications, Cambridge University Press, 2012

Doctoral Thesis

2012
Applications of Low-Rank Matrix Recovery Methods in Computer Vision

University of Illinois, Urbana-Champaign

Thesis Advisor: Prof. Yi Ma

Thesis Committee: (Late) Prof. Thomas Huang, Prof. Olgica Milenkovic, Prof. Sean Meyn

Journal Articles

2013
Fast ℓ1-Minimization Algorithms for Robust Face Recognition

A. Y. Yang, Z. Zhou, A. Ganesh, S. S. Sastry, Y. Ma

IEEE Transactions on Image Processing, vol. 22, pp. 3234–3246, 2013

Compressive Principal Component Pursuit

J. Wright, A. Ganesh, K. Min, Y. Ma

Information and Inference: A Journal of the IMA, vol. 2, pp. 32–68, 2013

2012
RASL: Robust Alignment by Sparse and Low-Rank Decomposition for Linearly Correlated Images

Y. Peng, A. Ganesh, J. Wright, W. Xu, Y. Ma

IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, pp. 2233–2246, 2012

TILT: Transform Invariant Low-Rank Textures

Z. Zhang, A. Ganesh, X. Liang, Y. Ma

International Journal of Computer Vision, vol. 99, pp. 1–24, 2012

2011
Toward a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representation

A. Wagner, J. Wright, A. Ganesh, Z. Zhou, H. Mobahi, Y. Ma

IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, pp. 372–386, 2011

2008
Robust Face Recognition via Sparse Representation

J. Wright, A. Y. Yang, A. Ganesh, S. S. Sastry, Y. Ma

IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, pp. 210–227, 2008

Conference Papers

2012
TILT: Transform Invariant Low-Rank Textures

Z. Zhang, X. Liang, A. Ganesh, Y. Ma

Asian Conference on Computer Vision (ACCV), 2010, pp. 314–328

Principal Component Pursuit with Reduced Linear Measurements

A. Ganesh, K. Min, J. Wright, Y. Ma

IEEE International Symposium on Information Theory (ISIT), 2012, pp. 1281–1285

2010
Fast ℓ1-Minimization Algorithms and an Application in Robust Face Recognition: A Review

A. Y. Yang, S. S. Sastry, A. Ganesh, Y. Ma

IEEE International Conference on Image Processing (ICIP), 2010, pp. 1849–1852

Robust Photometric Stereo via Low-Rank Matrix Completion and Recovery

L. Wu, A. Ganesh, B. Shi, Y. Matsushita, Y. Wang, Y. Ma

Asian Conference on Computer Vision (ACCV), 2010, pp. 703–717

Dense Error Correction for Low-Rank Matrices via Principal Component Pursuit

A. Ganesh, J. Wright, X. Li, E. J. Candes, Y. Ma

IEEE International Symposium on Information Theory (ISIT), 2010, pp. 1513–1517

2009
Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Matrices via Convex Optimization

J. Wright, A. Ganesh, S. Rao, Y. Peng, Y. Ma

Advances in Neural Information Processing Systems (NeurIPS), vol. 22, 2009

Fast Algorithms for Recovering a Corrupted Low-Rank Matrix

A. Ganesh, Z. Lin, J. Wright, L. Wu, M. Chen, Y. Ma

IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2009, pp. 213–216

Separation of a Subspace-Sparse Signal: Algorithms and Conditions

A. Ganesh, Z. Zhou, Y. Ma

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2009, pp. 3141–3144