Shivin Thukral

Machine Learning Engineer

Pinterest

About Me

I am a Machine Learning Engineer at Pinterest, where I work on building large scale recommendation systems for related pins using machine learning techniques on the Close Up Recommendations team.

I completed my MS in Computational Linguistics at the University of Washington in 2022 and graduated from BITS Pilani in Computer Science and Mathematics in 2020. Previously, I have had the opportunity to work at Avnio, Samsung Research Institute, and complete my thesis at the IMS Lab at University of Stuttgart.

Interests

• Natural Language Processing
• Machine Learning
• Computer Science

Education

• M.S. in Computational Linguistics, 2022

  University of Washington

• B.E. (Hons) in Computer Science, 2020

  BITS Pilani, K. K. Birla Goa Campus

• M.Sc. (Hons) in Mathematics, 2020

  BITS Pilani, K. K. Birla Goa Campus

Experience

 

 

 

 

 

Machine Learning Engineer

Pinterest

Jul 2022 – Present Palo Alto, CA

Working on building large scale recommendation systems for related pins

 

 

 

 

 

ML Engineer Consultant

Avnio

Sep 2020 – Sep 2021 Remote

Worked on NLP/ML use cases for parsing large RFP documents

 

 

 

 

 

Deep Learning Intern

HW Saver LLP

May 2020 – Dec 2020 Remote

Worked on developing seq2seq neural models for solving math word-problems

 

 

 

 

 

Intern

Samsung Research Institute

Jan 2020 – Jun 2020 New Delhi, India

Worked on development of robust Automatic Speech Verification (ASV) systems

 

 

 

 

 

Research Intern

Institute of Natural Language Processing, University of Stuttgart

Jul 2019 – Dec 2019 Stuttgart, Germany

Undergraduate thesis on the topic ‘Distributional Modelling of Cross-Cultural Judgement of Food Terms’

 

 

 

 

 

Summer Intern

Quorg Technologies

May 2017 – Jul 2017 Mumbai, India

Worked on App Store Optimization (ASO) and SEO techniques to increase ranking of app and website among search results

Publications

Multilingual Unsupervised Sequence Segmentation transfers to Extremely Low-resource Languages

We show that unsupervised sequence-segmentation performance can be transferred to extremely low-resource languages by pre-training a …

C.M. Downey, Shannon Drizin, Levon Haroutunian, Shivin Thukral

PDF Cite Preprint

Probing Language Models for Understanding of Temporal Expressions

We present three Natural Language Inference (NLI) challenge sets that can evaluate NLI models on their understanding of temporal …

Shivin Thukral, Kunal Kukreja, Christian Kavouras

PDF Cite Code Poster Slides Preprint

Detection of Malicious Android Applications: Classical Machine Learning vs. Deep Neural Network Integrated with Clustering

Today anti-malware community is facing challenges due to ever-increasing sophistication and volume of malware attacks developed by …

Hemant Rathore, Sanjay K. Sahay, Shivin Thukral, Mohit Sewak

PDF Cite Code DOI Preprint

Projects

*

All NLP Deep Learning Other

Distributional Modelling of Cross-Cultural Judgement of Food Terms

Undergraduate thesis completed at IMS Lab at University of Stuttgart

Slides

Robustness Testing of Neural Networks

Design project aimed at testing robustness of pre-trained classification models under adversarial noise attacks

Code

Emotion Analysis of Food Reviews

Project as part of the ‘Information Retrieval’ course to perform sentiment analysis of food reviews

PDF Code

Inventory Website Development for University Auditorium

Developed a website for tracking item inventory as per the requirement of Department of Backstage, BITS Goa

Code

Image De-Raining using Deconvolutional Networks

Project as part of the ‘Neural Networks’ course to perform de-raining of rainy images

PDF Code