Aditya Thyagarajan
Machine Learning Research Engineer at Cleanlab
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolAditya Thyagarajan is a Machine Learning Research Engineer with 11 years of experience building AI systems and improving data quality for real-world applications from Bengaluru. He combines strong academic credentials—a near-perfect MS in Computer Science from University of Colorado Boulder—with hands-on roles across startups and product teams, spanning speech noise reduction, NLP, and applied ML research. At Cleanlab he contributed substantive multi-label and class-wise thresholding improvements to a widely used open-source data-quality toolkit, showing a practical focus on messy, real-world labels. Comfortable bridging research and production, he has repeatedly moved novel algorithms into usable code and metrics-driven features. Colleagues describe him as detail-oriented and pragmatic, with a knack for refactoring complex codebases to unlock new capabilities.
11 years of coding experience
4 years of employment as a software developer
Bachelor of Engineering (BE), Computer Science and engineering, 8.88/10, Bachelor of Engineering (BE), Computer Science and engineering, 8.88/10 at M.S. Ramaiah Institute Of TechnologyMaster's degree, Computer Science, 3.97/4, Master's degree, Computer Science, 3.97/4 at University of Colorado BoulderSenior Secondary, Senior Secondary at National Public School
English, Kannada
Github Skills (8)
machine-learning10
data-quality10
python10
data-science10
numpy10
active-learning9
object-detection8
data-cleaning8
Programming languages (5)
Github contributions (5)
cleanlab/cleanlab
Oct 2022 - Jan 2023
The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
Role in this project:
Data Scientist
Data ScientistContributions:215 reviews, 7 commits, 29 PRs in 2 months
Contributions summary:Aditya contributed significantly to enhancing the multi-label classification capabilities within the `cleanlab` library. Their work involved refactoring and extending existing code to support multi-label classification tasks, as evidenced by the introduction of new helper functions and modifications to existing functions related to calculating label quality scores. The user also implemented crucial changes, such as those needed to support the identification of label issues, and introduced thresholding strategies for a better and class-wise performance in object detection problems.
operationdata-preprocessingtext-annotationreal-world-datadata-labeling
aditya1503/cleanlab
Sep 2022 - Feb 2024
The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
Contributions:421 pushes, 21 branches in 1 year 4 months
fairness-mlreal-world-datadata-integritydatacleaningdataframes
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.