Sanjay Saha

Sanjay Saha

ML Engineer @ ByteDance, PhD Candidate @ NUS

ByteDance, Singapore

National University of Singapore

Biography

I am a Machine Learning Engineer at ByteDance, currently working on Computer Vision problems. I am also a PhD Candidate at the Department of Computer Science, National University of Singapore. My research topics include analysis of synthesis of face images and videos, especially Deepfakes. You are welcome contact me via LinkedIn.

Download my resumé .

Interests
  • Computational Vision
  • Machine Learning
Education
  • PhD in Computer Science (awaiting defense), 2024

    National University of Singapore

  • Masters and BSc. in Computer Science and Engineering, 2014, 2016

    University of Dhaka

Experience

 
 
 
 
 
ByteDance
Machine Learning Engineer
Jul 2024 – Present Singapore
  • Work on Computer Vision and Multimodal Learning problems
  • Maintain services on the platform
 
 
 
 
 
National University of Singapore
Research Associate
Feb 2024 – Jul 2024 Singapore

Research on

  • Face Synthesis and detection
 
 
 
 
 
National University of Singapore
Research Assistant
May 2022 – Jan 2024 Singapore

Research on

  • Deepfake generation methods
  • Quantitative and Qualitative analyses
  • Reproduction and implementation of SOTA methods
 
 
 
 
 
National University of Singapore
Graduate Teaching Assistant
Aug 2018 – May 2022 Singapore

Conducted tutorial classes on (some multiple times)

  • Biometric Authentication
  • Advanced Analytics and ML
  • Workshop of Visual Computing, SCALE AI & ML
  • Others: Database Systems, Discrete Mathematics, and more.
 
 
 
 
 
Invigilo Technologies
Research Engineer (Part-time)
Jun 2020 – May 2022 Singapore

Developed Computer Vision solutions for automated construction work-site safety violation detection. Built solutions using state of the art video analytics techniques for

  • Detecting and tracking objects
  • Human action recognition
  • Pose estimation, proximity estimation, etc.
 
 
 
 
 
ByteDance
Research Scientist (Intern)
May 2021 – Jul 2021 Singapore
In this tenure I worked on a section of a project for TikTok video duplication detection where I had to work with variations of the transformer models, developing image matching algorithms based on requirements, working with SQL for informative report generation, etc.
 
 
 
 
 
University of Asia Pacific
Assistant Professor
Oct 2017 – Jul 2018 Dhaka, Bangladesh

Lectured, and designed curriculum of these courses

  • Structural Programming Language
  • Algorithm Design
  • Object Oriented Programming Also, guided two student-groups to research projects, one leading to a publication.
 
 
 
 
 
University of Asia Pacific
Assistant Professor
Oct 2017 – Jul 2018 Dhaka, Bangladesh

Lectured, and designed curriculum of these courses

  • Theory of Computation
  • Algorithm Design
  • Object Oriented Programming
  • Discrete Mathematics, Programming Language (Year 1) Conducted research on use of Machine Learning methods in different protein site identification (Bio-informatics) leading to multiple top-level journal publications.

Contact